Impact of Consumers ’ Moral Self-Identity on Green Purchase Decisions of Fast Moving Consumable Goods ( FMCG )

Much research has used general attitudes and behavioural models in terms of values, beliefs and norms as the predictors of green purchase intention. Green purchase intention refers to the willingness of a person to give preference to products having eco-friendly features in their purchase considerations. However, there is a dearth of studies that investigate this in relation to customers’ self-regulatory mindset. Hence, the present study aims to elucidate how customers’ moral selfidentity as a measure of regulatory perception guides individual decision making on green product purchase in Sri Lanka with special reference to green based Fast Moving Consumable Goods (FMCG). From extant literature, a conceptual model was formulated. It hypothesized that consumers’ moral self-identity is a positive predictor of their green purchase intention and in turn green purchase intention as a positive predictor of actual green purchase behaviour. A structured questionnaire was administered to a representative sample of 300 modern-trade retail customers. The data analysis comprised of linear regression analysis to test the hypotheses. The reliability of the measures was also tested. Customers’ moral selfidentity was found to be a significant positive predictor of green purchase intention for FMCG in Sri Lanka. Further, it was evident that green purchase intention is positively associated with actual green purchase behaviour. The findings have theoretical and practical implications for both academia and practitioners.


Introduction
In marketing theory and practice, "eco-friendly" or "green" has become a buzz word and a trend in recent years.Going green is a major contributor to the triple bottom line of business (i.e.people, profit and planet).However, going green requires a substantial shift in corporate culture as well as strategic decisions to be in line with consumers' environmental beliefs and attitudes.Concerns related to the environment are evident in the increasingly environmentally conscious market place.Majority of consumers have realized that their purchasing behaviour has a direct impact on many ecological problems (Han, Hsu, & Sheu, 2010;DiPietro, Cao, & Partlow, 2013).Customers adapt to this new threatening situation by considering environmental issues in their purchase decisions and buying only ecologically compatible products (e.g., products with biodegradable packages, Chlorofluorocarbon (CFC) free refrigerators and air-conditioners).
Studies on consumption of environmentally sustainable products have demonstrated that poorly perceived product performance is a significant barrier to their selection.Ottman and Reilly (1998) show that 41% of consumers do not buy green products because of their perceived inferiority.Since the 1980s, green marketing has gone through several stages.After a backlash in the 1980s, green marketing has made an upswing in the West from 2000 onwards (Ottman,, Stafford, & Hartman, 2006).The force of "going green" is now expanding to the Asian region, where environmental threats are alarming local governments and citizens.Like many Asian countries, Sri Lanka suffers from increasing levels of air toxic waste, exposure to severe road traffic noise, high levels of garbage deposal and rapidly lessening landfill space in most of the urban areas (Samarasinghe, 2012).
In order to advance a country's green revolution, McGougall (1993) asserts the essential role of consumers.This assertion is supported by the evidence that 30% to 40% of environmental degradation is brought about by the consumption activities of private households (Grunert, 1993).More importantly, if consumers exhibit a high degree of ecological concern and channel into some ecofriendly purchasing behaviour, it is likely that profit-driven enterprises will be strongly motivated to adopt the concept of green marketing in their business practices.
Consequently, the dynamics of this buyer-seller interaction will lead to further advancement of the green revolution across a country (Ottman, 1992).Thus, in order to better understand the environmental movement of a particular country, a practical approach is the examination of how the consumers in a country view ecological issues and how these views are reflected in purchasing behaviour with regard to green products and services.An important challenge that marketers face is to identify which factors lead consumers to buy green products or what consumers take into consideration before green purchases.If one can find out these factors, marketers will be able to develop effective marketing mix strategies to attract consumers towards green products and promote environmental sustainability.
With the increased level of environmental pollution and social issues, both consumers and business organizations are concerned with the natural environment in Sri Lanka.Consequently businesses have begun to modify their behaviour in an attempt to address environmental and social issues as a way of their social responsibility.Presently, it is observable that the Fast Moving Consumable Goods (FMCG) sector of the country is concerned about the green marketing issues.Most companies marketing FMCG have incorporated green elements into their marketing mixes as a competitive tool (Wanninayake & Randiwela, 2008).For instance, FMCG marketing companies in Sri Lanka use environment friendly and recyclable packaging and modify the products to minimize or prevent the environment pollution.Research evidence suggests that citizens in Asian societies are increasingly becoming sensitive towards environmental issues (Lee, 2008;Cherian & Jacob, 2012).Government policies and business strategies in many Asian countries are being reshaped to give more consideration to long-term sustainable development including environmental protection (Johri & Sahasakmontri, 1998;Lee, 2008).
Although environmental issues increasingly influence every aspect of human life as well as business strategies in the Sri Lankan context, still there are very few studies that discuss green issues in the Sri Lankan market context.Especially, there have been a few studies on green consumer purchase attitudes and behaviour undertaken in Sri Lanka.One survey method based study undertaken by Wanninayake and Randiwela (2008) investigated the consumer attitudes and perceptions towards eco-friendly products in FMCG sector and their willingness to pay for green products.Their study revealed that the Sri Lankan consumers have a substantial awareness of green products and they are willing to pay more for green products.Although the term "green products" is attracting a high attention of Sri Lankan people, the real factors affecting green purchasing decisions in the Sri Lankan context are not yet sufficiently discovered empirically.
Whereas the impact of environmental beliefs in terms of environmental knowledge and effect, as determinants of green purchase behaviour has been explained to a certain extent in Sri Lanka (e.g., D.S.R. Samarasinghe & Samarasinghe, 2010: D.S. Samarasinghe, 2012;G.D. Samarasinghe & Samarasinge,2013), consumers' general beliefs and knowledge about environmental aspect of products have been found to be poorly related to their green product purchase intention in Sri Lanka.Further, these studies highlight that future studies need to identify other socio-cultural and personal factors affecting green product purchase intention.Studies have suggested that that the purchase intentions of ethical consumers are driven by personal values (Chatzidakis, Hibbert & Smith, 2007) moral norms, internal ethics, and other similar factors (Arvola et al., 2008;Shaw & Shui, 2002;Vermeir & Verbeke, 2008).
In this context, the researchers find that there is a need to identify a person's ethical or moral selfidentity as an important antecedent for their green purchase intention in Sri Lanka.self-identity is an important part of an individual's self that relates to a particular behaviour (Conner & Armitage, 1998).Meanwhile Shaw, Shiu, and Clarke (2000) recognized that ethical motives may become part of consumers' selfidentity.Literature on moral of a person, () identifies that consumers who have a highly self-important moral identity are more likely to show cooperative behaviour, social responsibility as well as self-expressiveness (Reed, Aquino & Levy, 2007).Literature identifies that an ethical consumer is ecologically conscious and favors products that are eco-friendly and less harmful to the society (Harper & Makatouni, 2002).The study of Honkanen, Verplanken and Olsen (2006) has established that ethical motives are related to positive attitude and purchase intention towards ethical consumption.Thus, the present study formulates research issue as to how moral selfidentity of consumers influences their green purchase intention and actual green behaviour.
The primary aim of the study is to elucidate the linkage between Sri Lankan consumers' moral self-identity and their green purchase behaviour in relation to FMCGs available at modern trade retail chains.Thus, it mainly focuses on achieving the following objectives: 1.To explain the relationship between moral self-identity and green purchase intention.2. To elucidate the link between green purchase intention and actual green purchase.
In the following section, the paper presents key literature on green consumerism, theoretical models on consumer attitudes and behaviour, moral self-identity and linkage between, moral selfidentity, green consumer purchase decision and actual green behaviour.It then develops main hypotheses and the conceptual model of the study.Next, it presents methods adopted for the study, discussion of results and finally conclusion and implications.

Green Consumerism
The intensification of the environmental crisis has reached alarming extents, which, for some, has fostered a greening of public perception (Babiak & Trendafilova, 2011;Kouchaki, Smith-Crowe, Brief & Sousa., 2013).For example, a niche market of ecologically minded consumers has emerged, provoking a variety of corporate responses to this popular reevaluation of human consumption habits.Mei, Ling and Piew (2012) citing the study of Renfro (2010) ascertains that green consumers as the consumers who support businesses that operate in environmental friendly ways.In addition, green consumers are also concerned about how green are the products that they purchase.For example, healthconscious consumers purchase products that are organic, nontoxic, water-based, and natural (Ottman & Reilly, 1998).
Green consumerism is on the rise in America; it has been noted that more people recycle than vote for president.The environmental effects of green consumerism are contested.Some argue that it contributes to the greening of American consciousness, in part because it creates a balance between consumer's expectations and business profit motives (Kouchaki et al., 2013).However, others argue that green consumerism does not address the root cause of environmental problems because it does not encourage a decline in consumption, and in fact encourages corporate green washing that misleads consumers about "sustainable" business practices (Stauber & Rampton, 1995;Kouchaki et al., 2013).Empirical evidence shows that although increasing numbers of consumers have absorbed and are motivated by the values of green consumerism, a change in consumption behaviour is much less apparent (Carrington, Neville, & Whitwell, 2010).This emphasizes the need for research to explain the gap between what ethically consumers intend to do and what they actually do at the point of purchase (Shaw, Shiu, Hassan, Bekin & Hogg., 2007).
While green marketing has a variety of connotations, it can be broadly understood as commercial messages and consumer behaviour that reflect "concern about the effects of manufacturing and consumption on the natural environment" (Wagner, 1997, p.1). Green consumers share common values in their concern for the state of the environment and their awareness of the effect of overconsumption on the rate of environmental devastation.These values are reflected in an increased emphasis on the eco-costs of a product, which include its collective social and long term economic impacts.Sustainable marketing conceptualizes production-consumption systems with several basic tenets: ecosystems pose a physical limitation on production; businesses must account for the entire product life cycle; pollution prevention and resource recovery are appropriate sustainable business practices; small environmental improvements translating into large absolute improvements; and sustainable marketing is not an exercise in corporate altruism, but a social obligation (Fuller 1999).These indicate the unique set of responsibilities that environmental marketers encounter.
Green consumerism is one main aspect of ethical consumerism.Ethical issues comprising of environmentalism are incorporated within ethical consumerism in which there are complex decision making processes for ethically conscious consumers (Freestone & McGoldrick, 2008).Further, green consumer goods are ideological by their very nature, and thus represent an ethics-based market with a consumer culture shaped by environmentally aware shoppers.Thus, eco-marketing constructs a complicated ethical identity for the green consumer.Environmentally aware individuals are already guided by their personal ethics.In trying to attract new consumers, environmentally-minded businesses attach an aesthetic quality to environmental goods.

The Theory of Reasoned Action and the Theory of Planned Behaviour
The theory of reasoned action (TRA) (Fishbein & Ajzen, 1975;Ajzen & Fishbein, 1980) is a theory of attitude-behaviour relationships which links attitudes, subjective norms, behavioural intentions and behaviour in a fixed causal sequence.'Behaviour is postulated as a result of behavioural intention, which in turn is postulated 'to arise out of a combination of people's attitude towards performing the behaviour in question and their perceptions of the social pressure put upon them to perform that behaviour' (Sparks & Shepherd, 1992, p.388).Attitudes are said to arise from a combination of people's beliefs about behavioural outcomes and their evaluations of those outcomes.This latter facet of the theory has clear underpinnings in subjective expected utility (SEU) theory (Ajzen & Fishbein, 1972).Subjective norms are suggested as having similar origins in a combination of people's perceptions, on what important others think the individual should or should not perform The theory of TRA of Ajzen and Fishbein (1972) was applied to study volitional behaviours and has now been applied widely in a number of domains (e.g., Ajzen & Fishbein, 1980).Later, Ajzen (1985) has introduced the "theory of planned behaviour" (TPB), adding a measure of perceived control to the basic structure of the TRA.This modification has been advanced in order to extend the domains of behaviour covered by the TRA to behaviours that are not totally under a person's control.Empirical applications of this "extended" model have appeared subsequently in the literature.An attitudesbehaviour gap exists within the ethical consumerism literature pertaining to the purchase decisionmaking process (Carrington et al., 2010).Application of the TPB within the ethical or green consumerism context has concerned the actual control the individuals have over their personal behaviour at the point of purchase and how this differs according to their own perceptions of behavioural control when they are formulating their purchase intentions.Thus, it is important to incorporate ethical or moral self-identity of the consumer in a green context in exampling purchase intention.

Role of Self-Identity within the Theory of Reasoned Action
A person's self-identity (or self-concept) has been viewed as an important influence on behaviour (Markus, 1980;Rosenberg, 1981;Turner, 1982).The precise use of terminology and the meaning of "identity" vary both between and within disciplines (Gleason, 1983).Despite some divergence of interest between the sociological emphases on social roles and the psychological orientation towards the study of "personality" characteristics, a concern with self-identity issues is nevertheless apparent in many areas of both disciplines.The recent claim that self-identity will contribute to the prediction of behavioural intentions independent of how attitudes are made, is illustrated empirically, in the works of Biddle, Bank and Slavings (1987) and Charng, Piliavin, and Callero (1988).Evidence for the effect of self-identity on behaviour, independent of behavioural intentions is provided by Granberg and Holmberg (1990).All these studies significantly identify the role of self-identity within the structure of the TRA.
Following Charng et al., (1988), Granberg and Holmberg (1990), a study of voting intentions found that both self-identity and prior behaviour had independent effects on behavioural intention and voting behaviour.No measures of attitudes or subjective norms were taken in their study.Biddle et al., (1987) proposed that different "intellectual traditions" of self-concept and attitude perspectives in the literature may be partly responsible for the lack of integration of the two orientations.Although this statement probably contains some truth, it is also likely that attitude researchers aligned with the theoretical position of Fishbein and Ajzen (1975) would consider that a person's self-identity is reflected in that person's beliefs, values, and attitudes and that the inclusion of self-identity as a component of a model should not represent a theoretical or an empirical advancement.
Although behaviour researchers consider that a person's self-identity is conceptually distinct from his or her evaluative attitudes, a bi-directional causal link is likely to exist between a person's selfidentity and his attitudinal evaluations.If a person's self-identity per se is important to him or her (as suggested by Biddle et al., 1987), then the creation, affirmation, or bolstering of self-identity through the performance of behaviour could be conceptualized validly as an outcome of the behaviour, albeit possibly difficult to articulate and less salient in relation to other more tangible outcomes.Hence, selfidentity is needed to position as one important determinant that predicts behavioural intention.

Moral Self-Identity as an Extension to the Self-Identity
Aquino and Reed (2002) proposed that people possess a cognitive schema of the moral self that is organized around a set of moral trait associations.Aquino and Reed's (2002) introduced a trait based definition of moral identity built on the work of other writers such as Brewer and Gardner (1996) who argue that traits are the loci of self-definition around which personal identities are structured.The study adopts this definition of moral identity as a mental representation (i.e., a self-image) that a consumer may hold about his or her moral character (Aquino & Reed, 2002).For example, a consumer may possess in his or her mind a typical exemplar of what a moral person is like in terms of characteristics (e.g., being kind, honest, and loving), feelings (e.g., showing concern or empathy for others), and behaviours (e.g., helping strangers, engaging in charitable behaviours).A consumer's moral identity (the extent to which this image is an important part of his or her self-concept) may motivate choices and the pursuit of actions that demonstrate social responsiveness to the needs of others as cited in Reed et al. (2007).Aquino and Reed (2002) citing the studies (Erikson, 1964;Blasi, 1984;Damon & Hart, 1992;Hart, et al., 1998) stress moral identity as one kind of selfregulatory mechanism that motivates moral action.

Green Purchase Intention vs. Actual Green Purchase Behaviour
In marketing and consumer behaviour literature, it has become very important to predict the actual purchase behaviour of customers.However, due to the practical difficulty in measuring the actual behaviour, the purchase intention has been used as the most immediate proxy to predict the actual behaviour (Ajzen & Fishhbein, 1977;1980).Green purchase intention is defined as the probability and willingness of a person to give preference to products having eco-friendly features over other traditional products in their purchase considerations (Rashid, 2009).Chan and Lau (2000) defined green purchase as a specific kind of eco-friendly behaviour that consumers perform to express their concern to environment.Purchase intention is a critical factor to predict consumer behaviour (Fishbein & Ajzen, 1975).
According to the theoretical framework of Ajzen and Fishhbein (1980) the study defines green behavioural intention as the purpose towards purchasing green brands / products and intention to involve in social and community activities to protect the environment whilst the actual green behaviour is referred to the degree of real purchase of the green products and present involvement of social, legal, political and environmental friendly works to protect the environment.Studies of Shaw & Clarke (1999) and Shaw & Shui (2002) have tended to accept the theoretical assumption that an individual's intentions will directly determine their actual behaviour (Fukukawa, 2003).This assumption, however, has been widely criticized as an oversimplification of the complex transition from intentions to action (Bagozzi, 2000;Morwitz, Steckel & Gupta., 2007).Furthermore, empirical studies in the field of consumer behaviour more broadly suggest that purchase intentions do not translate literally into purchase behaviour (Morwitz et al., 2007;Young, De Sarbo, & Morwitz., 1998).

Hypotheses of the Study
Green Purchase behaviour is a morally motivated action.Reed et al. (2007) reported that Damonand Hart (1992) argue that: There are both theoretical and empirical reasons to believe that the centrality of morality to self may be the single most powerful determiner of concordance between moral judgment and conduct.People whose self-concept is organized around their moral beliefs are highly likely to translate those beliefs into action consistently throughout their lives.(p.455) For some consumers, moral identity is both the antecedent and the outcome of the need for selfconsistency between who a person thinks he or she is and how he or she acts in the world (Blasi, 1993).Based on Aquino & Reed's (2002) argument that moral identity as one kind of self-regulatory mechanism that motivates moral action, green purchase decisions of customers can also be recognized as a morally motivated action in literature.Arvola et al., (2008) included moral norms to predict purchase intentions of organic food meanwhile Vermier and Verbeke ( 2008) integrated the role of personal values within the purchase intentions of sustainable food.In line with this argument, consumer's moral self-identity can theoretically be an effective contributor to identify target markets and motivate consumers to engage in green purchasing behaviour.Thus, this empirical study hypothesized that; H1: Consumer's moral self-identity will positively influence purchase intention for green products.
Literature demonstrates a positive relationship between environmental purchase intention and behaviour (G.D. Samarasinghe & Samarasinghe, 2013;Zhu, Geng & Qi 2013).Most of the empirical studies used the "theory of reason action" which describes that the intention is the most immediately relevant predictor of corresponding behaviour (Ajzen & Fishhbein, 1977;1980).Although above mentioned theories attempt to explain the gap between consumer's attitudes and behaviour, still there is doubt in different empirical domains that these attitudes do not always results in actual behaviour.Based on the arguments of Hines, Hungerford and Tomera (1987); Steg (2005) andDittmar's theory (1992), it develops the link that people develop motivation and intention to buy green products leading ultimately to actual green purchase behaviour for them.Thus, it is hypothesized that; H2: Consumer's green purchase intention will positively influence actual purchase of green products.
Based on the above literature review, the following conceptual model (shown in Figure : 01) was developed to show the hypothesized relationships between the independent variables and the dependent variable of the study.This model assumes that moral self-identity of consumers, as a predictor of green related purchase behaviour, would automatically result in more pro-environmental behaviour.Literature on moralselfidentity has supported that consumers with more need for moral self-identity are concerned about the environmental problems and are inclined towards green purchase (Damon & Hart, 1992;Blasi, 1993).

Methods
Following the positivistic research tradition and the quantitative research approach, survey research strategy was adopted.The population of the study consisted of local customers attending modern trade / supermarket outlets in the Colombo district of Sri Lanka.It mainly focused on ecofriendly fast moving consumable products available in modern trade/ supermarket outlets.

Operationalization
Moral self-identity of consumers was operationalized using Aquino and Reed's (2002) trait based two dimensional instrument, which consists namely symbolization and internalization.Symbolization includes six items tapping the degree to which the traits are reflected in the respondent's actions in the world whilst internalization consists of five items tapping the degree to which the moral traits are central to the self-concept.Green purchase intention was measured in terms of the respondent's intention to engage in green purchase using four items whilst actual green purchase behaviour was measured in terms of the respondent's actual present behaviour with respect to green purchase using two statements (Li, 1997).

Data Collection & Analysis
Self-Administrative Questionnaires were used to collect the predetermined data required to measure the constructs related to green purchase behaviour.A structured questionnaire mainly consisting of 5 point -Likert scales (Strongly Disagree -Strongly Agree) ranging 1-5 was used to measure the constructs.In order to make inferences of the population, it was determined that a sample of 300 consumers would be sufficient for the study as it satisfies the rule of thumb proposed by Roscoe, 1975(cited in Sekaran, 2007, p.295).In order to collect the data necessary to measure the constructs related to the subjects of the study, it was required to interview customers attending the respective supermarket chains in Colombo district of Sri Lanka.A mix of convenience and judgemental sampling techniques was employed in selecting the respondents attending supermarkets in the Colombo district of Sri Lanka.It was made sure that the respondents covered basic demographics such as gender and various age groups.The response rate was around 80% as few respondents were not willing to respond the questionnaire (i.e., 237 respondents completed the questionnaires).Data analysis in the study consisted of both descriptive statistical analysis and inferential statistical analysis.The hypotheses one and two were tested using multiple linear regression analysis.The Statistical Package for Social Sciences (SPSS R ) was used for data analysis.

Green Purchase Intention
Actual Green Purchase Behavior H1 H2

Respondents' Demographic Profiles
Based on this survey, the male respondents represented 46 % (109 male respondents) of the total respondents while the female respondents represented 54 % (128 female respondents) of the total respondents.Further, the age distribution of the sampled respondents were between the age of 20-30 years old (29.4 %) scored the highest respondent's age group, followed by between 31-40 years old (25.1 %); between 41-50 years old (19.4 %); over 50 years old (16.1 %) and the lowest score of the respondent's age group is age less than 20 years old (10 %).

Reliability and Validity of Measures
Before conducting the large scale questionnaire survey, a pilot study was carried out to confirm the reliability of the constructs.Cronbach's alpha was calculated to measure the reliability / internal consistency of the measurement scales.The overall construct of moral self-identity (11 items) had a Cronbach's alpha of 0.74; green purchase intention (four items) had a Cronbach's alpa of 0.81 whilst actual green purchase behaviour (two items) received an inter-item correlation of 0.88.As all three constructs scored the Cronbach's alpha or correlation values above 0.7, all scales were sufficiently reliable (Nunnally, 1967).The face / content validity of the scales was assured through experts' reviews and literature survey.In order to test the dimensionality of the measurement constructs, an exploratory factor analysis was performed by following the procedure recommended by Churchill (1979).
All of the data measuring independent and dependent variables were obtained from the same source (i.e.customers) using the same method.The issue arises therefore as to whether the covariance between the constructs is an artifact of single-source common method bias.To address this issue, Harman's single-factor test was conducted on all of the items used to test the hypotheses that were obtained from customers, (moral self-identity, green purchase intention and actual green behaviour).The results from this analysis revealed that the first factor accounted for 32 per cent of the total variance in the items, which indicates that common source/method variance does not explain the majority of the covariance between the scales (P.M. Podsakoff, MacKenzie, Lee, & Podsakoff., 2003).

Association between Key Constructs
According to the Pearson correlation analysis, it was evident that moral self-identity had a significantly moderate and positive association (r=0.68,p=0.000) with green purchase intention, while green purchase intention had a significantly strong positive association (r=0.82,p=0.000) with actual green purchase behaviour.

Test of Hypotheses
In testing the two hypothesized relationships in the conceptual model, the linear regression analysis revealed the following results as shown in the Table 1.

Table 1: Linear Regression Results
The results of this analysis are given in the table: 1.According to this analysis, hypothesis-one is accepted.There is significant evidence (Beta = 0.58, p = 0.007) to conclude that the consumer's moral self-identity positively influences the green purchase intention for FMCGs in Sri Lanka.This is particularly because customers who are most likely to reflect moral and ethical selves may be more likely to express moral action that shows social responsiveness to the needs of others (Kohlberg, 1969;Rest, 1980;Gilligan, 1982).This finding is consistent with the extant literature and research undertaken in the West where customers' moral self-identity is a powerful motivator of their purchase decision making and guiding their intention for environmental friendly products and services (Blasi,1993;Aquino & Reed, 2002;Reed et al., 2007).This confirms that customers' positive feelings and emotions related to their need for presenting themselves as a moral valued personality are related to purchase of environment friendly products and services (Arvola et al., 2008;Vermier & Verbeke, 2008).This can be a strong motivator for creating a purchase intention towards them in the Asian context as in the West.
Further, hypothesis-two is accepted.There is significant evidence (Beta= 0.72, p=0.004) to support that green purchase intention leads to actual green purchase decision in Sri Lanka as in the West.This confirms the claims of the seminal research papers (Ajzen & Fishhbein, 1977;1980).Hence, purchase intention for green products is still a powerful predicator for actual green product behaviour in the Eastern context, especially in Sri Lanka.This finding is also consistent with the past studies of G.D. Samarasinghe and Samarasinghe (2013) and Zhu et al., (2013).
In the regression models for predicting purchase intention and actual green purchase respectively, the goodness of fit is predicted using R 2 values which are respectively 0.46 and 0.67.Therefore, the model predicting green purchase intention using moral self-identity explains only 46% of the total variance of green purchase intention.Meanwhile the regression predicting actual green purchase behaviour using green purchase intention explains only 67% of the total variance of the actual green purchase behaviour.
In further analysis of relationships, it is interestingly found that the relationship between moral self-identity and actual green purchase behaviour is partially intervened by green purchase intention in Sri Lanka according to the hierarchical regression analysis procedure recommended by Baron and Kenny (1986) to test the mediator relationship.This finding is also consistent with the Western empirical studies that used the "theory of reason action" that intention is the most immediately relevant predictor of corresponding behaviour (Ajzen & Fishhbein, 1977;1980) as well as the findings of the Sri Lankan study of G.D. Samarasinghe and Samarasinghe (2013).

Conclusion and Implications
Based on the above empirical findings, it is evident that in the Sri Lankan context moral selfidentity of consumers is a significant predictor of green purchase intention.This finding is consistent with the studies of Arvola et al. (2008) andVermier &Verbeke (2008) that incorporated moral norms to predict purchase intention for eco-friendly products.This might be an indication that rather than general green beliefs and attitudes, moral and ethical value based self-identity as a personality related dimension, can affect purchase intention towards environmental friendly product purchase decisions of the Sri Lankan consumer.This might have been influenced by the religious and spiritual values inherent in the Sri Lankan culture as well as the experiences of global environmental issues and mind set of the people of the country.

Managerial and Theoretical Implications
These findings have managerial implications for Sri Lankan marketers to develop eco-friendly brands and packages that target the moral self-important customer segments in Sri Lanka.FMCG marketers can incorporate more ethical criteria and highlight moral aspects in their product and promotion decisions to differentiate their brands.Thus, customers with more need for moral identity can be motivated to buy these brands.Further, eco-labeling and promotional materials can be designed more attractively to get these customers attending modern trade outlets.
Theoretically, the present body of academic literature has noted that there are some discrepancies between attitudes and behaviour gap related to pro-environmental behaviour due to differences in people's experiences, cultural traditions, norms and research methodological problems etc.As a response to this, the present study reveals that moral or ethical self-identity of customers can be a common predictor for ethical and green based product purchase decisions in both Asian countries which are said to have more of a shame culture and Western countries which are based on more of a guilt culture.Further, moral self-identity of customers can be used as a more specific attitudinal measure to predict green purchasing behaviour instead of using general attitudes and beliefs related measures available in traditional consumer behaviour literature.

Future Research Opportunities
The findings highlight the need for future studies that should explore factors other than moral self-identity that guide green purchase intention in a developing context like Sri Lanka.The moderate level of predictive power of moral self-identity as the only independent variable in the model motivates future studies to include other unique predictors of green purchase intention such as social class classifications (e.g., new emerging middle class of Sri Lanka as proposed by Liyanage, 2003) and various other demographic factors such as age, income, occupation education of Sri Lankan consumers.The study measured moral self-identity in the light of internalization and symbolization.Deviating from this trait based operationalization, it is also of paramount importance to identify more precise variables and aspects of moral self-identity and develop better measures in the Sri Lankan context.Further, it would be interesting to include religiosity and personality related variables together with moral-self-identity as predictors of green purchase intention in future studies.As a controlling variable, the impact of perceived price, quality and brand reputation of FMCGs can also be introduced in the conceptual model and measured, to make the findings more conclusive and robust.Furthermore, it is imperative to measure actual green purchase behaviour using more reliable methods such as observations, as the present study employed survey based questions to measure the actual green purchase behaviour.Finally, the finding that explains the moderate predictive power of green purchase intention as a predictor of actual green purchase behaviour leaves room for future researchers to identify other significant predictors of actual green purchase behaviour in Sri Lanka.

Figure
Figure 1: Conceptual Model