Religiosity and Intention towards Saving of a Muslim Minority in Greece

Culture influences financial decisions of individuals therefore, by examining the cultural background of cultural differences stemming from difference in religious beliefs is of significance to better understand their preferences and financial decisions. This research examines the relation of religiosity and intention towards saving of Pomak households, a Muslim minority living in the northeaster part of Greece. Quantitative data, collected with field research using the Religious Commitment Inventory (RCI-10) and Warneryd’s saving intention scale was used. The results reveal that there is a strong relation between religiosity and intention towards saving, indicating that the intention to save is affected by the religious norms of a person.


Introduction
In a constantly changing and increasingly globalised world, religions still play a significant role in influencing social and saving behaviour. However, although researchers have long recognised the importance of religious value systems in sociology (Anderson, 1970) and in psychology (Allport, 1967;Pargament & Hahn, 1986), also, many authors such as (Arrunada, 2010;Belzen, 1999;Guiso et al., 2003;Karijin et al., 2007;McCleary, 2007;Mokhlis, 2009;Renneboog & Spaenjers, 2012) have admitted and exhibited in full the importance of the religious sentiment in studies that examined their saving behaviour. The aim of the present investigation is to explore the effects of religion beliefs on the intention of a person towards saving in the Muslim minority of Pomaks in Greece. The Pomak population is living in the Greek part of the northeast region of Thrace, in three different prefectures: Xanthi, Rodopi and Evros. The sub goals of this research are to explore the religiosity of the sample according to their socio-demographic profile and to examine the differences between high and low religious individuals on the intention towards saving. The understanding of the saving behaviour of this minority living in this region is of a great interest for financial institutions and economic authorities. Minority populations are mainly placed in a disadvantageous situation when they apply for a loan, either from a bank or a similar economic establishment (Blanchard et al., 2008). Specifically, for Pomak minority, the socioeconomic segregation was the main reason why the moved their funds to other countries (mainly Turkey), at least until the early '90s when restrictive measures prevented them from investing them in the region (Anagnostou, 2003). Once these measures were abolished in 1991 the situation changed again (Anagnostou, 2003;2001). The research also provides new insights into the relative emphasis of the various factors that influence saving behaviour in present days. The article is structured as follows. In section 2 there is a short recounting of the literature used for religiosity and intention towards saving, section 3 refers to the method and the statistical analysis which were used. In section 4 involved the outcomes and analysis of data, while section 5 presents the final conclusions. security for any, predictable or not, future events and enhances their living standards (Anong & DeVaney, 2010;Hira, 1987;Lee et al., 2000). Such decisions have a profound impact on how the household financially performs. There are essential reasons why economists and researchers are so deeply concerned with the way religious institutions function and how that affects the economy. Undoubtedly, one reason is the degree to which religion affects the formation of cultural conceptions in every society. This fact renders religion fascinating as a field of study, as it can and indeed does influence individuals as well as whole classes, groups or nations in terms of behaviour, sets of values, beliefs and ways of acting (Khraim, 2010). Religiosity is defined as "the extent to which an individual is committed to the religion he or she professes and its teachings, such as the individual's attitudes and behaviours reflect this commitment" (Johnson et al., 2001, p. 25). Being religious acquires different meaning for different persons (Slater et al., 2001) thus, making it more difficult for social scientists and marketing researchers to analyse. Religiosity is important as it is capable of influencing an individual cognitively and behaviourally. Religious persons have value systems that differ from those of the less religious and the non-religious (Hifza & Hashanah, 2015). In accordance with Islam, one's convictions and submissiveness must be evident in the way they act. Therefore, Muslims should not just uphold their beliefs, but practise them, as well; otherwise, being forgetful or disobedient is considered a sin, a non-conformity to their religion (Souiden & Rani, 2015). This set of principles that renders the divine in the centre of their existence, is supposed to act as a constant reminder and guide of how their whole life must adapt and evolve in order to fall in step with their religion (Souiden & Rani, 2015).
As Muslims are guided by Islam in most aspects of their lives, its part in their handling of finances has been widely studied, as well. Delener (1994) concludes that there are four division of religiosity that people fit into: the highly religious, the moderate, the not so religious ones and not religious at all. Those levels appear to be related to their economic attitudes; those who are less religious don't seem to be influenced in one way or another when it comes to making financial decisions.
Religion commands that people stick to certain rules as far as their dress code, eating habits, ideas and spending are concerned (Souiden & Rani, 2015). Mokhlis (2009) concludes that Muslim purchases are highly affected by their Islamic beliefs when it comes to shopping cars, food and clothes in Malaysia. Bonne et al. (2007) researched the purchase of halal meat in France and found out that the spending behaviour of Muslims is not similar to that of non-Muslims. Shaari & Afirin (2010) reached the same conclusion and underlined the role of religion in individual lifestyles. Khraim (2010) also affirms this finding. Still, there should be a dividing line drawn between what Muslims believe and how they really behave; meaning, there are examples of Muslim people who, in spite of their religion, tend to indulge into a different lifestyle. According to previous studies (Arrunada, 2010;Belzen, 1999;Guiso et al., 2003;Karijin et al., 2007;McCleary, 2007;Mokhlis, 2009;Renneboog & Spaenjers, 2012), intention towards saving have found to be significantly differentiated among the group of lower and higher religiosity. In contrast, other studies have shown that religiosity is not an important factor of intention (Dusuki & Abdullah, 2007;Erol et al., 1990;Gerrard & Cunningham, 1997;Haque et al., 2009). Understanding religiosity would allow to advance in this point, and to go deeper into uncovering the correlation between it and saving performance (Akerlof & Kranton, 2011), regarding the Pomak minority in Greece.

Data & Methodology
A survey of Pomak households was conducted to collect data anonymously and on voluntary basis. A self-completion survey was given to 600 respondents in the region of Thrace in three different prefectures of Xanthi, Rodopi and Evros. The adult head of the household was questioned in order to retrieve information for the household. Households in the survey were selected through the snowball sampling procedure. From the total of the participants, 410 were male (68%) and 190 were female (32%). Most of the participants were living in the prefecture of Xanthi (N = 386, 64.3%), while 183 (30.5%) were living in Rodopi and 31 (5.2%) in Evros.
The Religious Commitment Inventory-10 (RCI-10) of Worthington et al. (2003) was used. The RCI-10 is a brief 10-item screening assessment of the level of one's religious commitment using a 5-point Likert rating scale from 1 (not at all true of me) to 5 (totally true of me). Religious commitment refers to how much an individual is involved in his or her religion (Koenig et al., 2012). In particular, a religiously committed person is supposed to "adhere to his or her religious values, beliefs, and practices and use them in daily living" (Worthington et al., 2003, p.85). RCI-10 examines intrapersonal religious commitment (6 items) and interpersonal commitment (4 items) levels of the individual. An example of typical items in scale includes, "I enjoy working in the activities of my religious organization" and from the interpersonal subscale "I enjoy spending time with others of my religious affiliation".
For measuring the intention towards saving the Warneryd's saving intention scale was used, which consists of twenty-one attitude questions describing the individual's intention to saving (Warneryd, 1996a;1996b). He ach.ccsenet.org Asian Culture and History Vol. 12, No. 1;2020 introduced five subscales named "thrift", "saving involvement", "saving habits", "shame of debt", and "no need to save". According to Warneryd (1999, p.73) thrift is "a summary concept for all sorts of psychological factors that were assumed to affect savings". The saving involvement subscale depicts how each person conceives and renders significance and concern on any given item in his environment (Zaichkowsky, 1986). The saving habits subscale reflects the social comparison effect that essentially influences the saving behaviour (Duesenberry, 1949;Fisher, 1930;Keynes, 1936). Shame of debt subscale demonstrates the unanticipated events that certainly contribute to debt perceptions of the individuals' future needs and prompt their actions (Warneryd, 1999). Finally, no need to save subscale represents the individual's perception that living on less will pay off in the future (Kouzes & Posner, 2011). Those who are more aware of ways to save tend to act differently as compared to those who aren't (Bernheim & Garrett, 2003;Lusardi, 2008).
Pearson correlation coefficients were computed to quantify the linear correlation of the scales of religiosity and intention towards saving. The descriptive K -cluster method was applied to split the sample in two homogeneous groups regarding to the religiosity and intention towards saving scores. Student t -test for two independent samples was conducted to evaluate the statistical significance of the difference among the two clusters that were defined after the K -clusters method. As the sample was a sample of convenience, it was not reliable to use a statistical method such as multinomial probit, which statistically generates conclusions from the sample for the population. Instead, the K-cluster descriptive method was chosen which is simple in both the application and the presentation of the findings.

Results
Religiosity was found to be positively correlated to intention towards saving (Table 1). In particular, all intention subscales was significantly positively correlated with religiosity, with the subscale of thrift having the larger correlation (r(600) = 0.451, p < .01) while on the other hand shame of debt had the smaller coefficient (r(600) = 0.087, p < .05). *Correlation is significant at the 0.05 level (2-tailed).

Source: Authors' calculations
The above results suggest that there are specific differences in the religiosity of the sample according to their socio-demographic profile. In order to enlighten and clarify the differences between high and low religious people, the K -cluster method was applied to separate the sample into 2 groups according to their religiosity and their intention toward saving subscales ( Table 2). The first group (N = 292) was the group of lower religiosity (M = 23.5, SD = 5.7) while the second (N = 308) was the group of larger religiosity (M = 36.9, SD = 4.7). The lower religiosity group was the younger part of the sample (M 1 = 28.7, SD 1 = 10.6 vs M 2 = 36.9, SD 2 = 4.7), being single or married without children (Group1: 69% vs Group2: 55%) and more educated that the second group (percent of high school graduation Group1: 77% vs Group2: 62%). The religiosity scores of the Group1 was comparable to previously reported findings of group of general population in various religious or minority groups (Table 4, row 1, 2, 3, 15) whereas the Group 2 was comparable to persons having a strong religious background (Table 4, row 4, 5, 12, 14).  Thus, the above findings suggest that the level of the religiosity of a person is strongly related to other important decisions that this person is taking in his/her life. It is worth noting that the effect of religiosity on intention at the high religiosity group was significantly larger of the corresponding effect on the first group (Table 3), indicating that the intention to save for the younger and better educated Pomaks is not highly affected by the Muslim cultural tradition.
Specifically, Table 4 summarises the means and standard deviations for groups totalling almost 2000 people. For secular groups, means for the full-scale RCI are between 21 and 26 (SDs between 10 and 12). According to theorizing and reviews of research (Worthington et al., 1996;Worthington, 1988), people scoring greater than one SD higher than the mean should be considered highly religious. There is evidence that they view the world more in terms of their religious worldview than do less religious people or nonreligious people. Therefore, religion is expected to play a part in counselling and certainly in the perception of clients. Thus, according to theory (Worthington, 1988), a full-scale RCI-10 score of 38 or higher would justify considering a person to be highly religious. Source: Worthington et al. (1996;1988)

Conclusions
This research offers substantial theoretical value, which helps better comprehend the importance of religiosity that is significant as a cultural-based predictor of the Muslim's saving behaviour and it could prove to be a valuable resource to the financial institutions in order to discover new opportunities for considering this minority group as part of their customers. According to the results of the research the effect of religiosity on intention at the high religiosity group was significantly larger than the corresponding effect on the lower religiosity group, indicating that the intention to save for the younger and better educated Pomaks is not highly affected by the Muslim cultural tradition. Young Pomaks savers consist an intriguing mass of people for marketing researchers, as these attempts to form their identity and attitude on the basis of how they save and spend money. This pattern of behaviour defines who they are (Holbrook & Schindler, 1989). They can be strong influencing forces for their surroundings, in terms of marketing interest, and forge other behaviour on financial issues (Grant & Waite, 2003). Young-adult consumers act as agents of change within their social and cultural circles (Leslie et al., 2001). Therefore, young Pomaks are depicted as an intriguing group with regard to international marketing.
In an international setting, concerning financial institutions, the present research can be viewed as an effort towards the identification of religiosity as a factor influencing individual's saving behaviour. Previous studies have thoroughly documented the importance of religious factors affecting bank customers' attitudes and saving preferences (Baele et al., 2014;Dahari et al., 2015;Metwally, 1996;Newaz et al., 2016;Souiden & Rani, 2015). Thus, there seems to be also an acquisition of useful information for both economists and sociologists on the saving behaviour of the Pomaks. Therefore, the financial institutions can develop financial products or services according to their needs and preferences so as to match or exceed their expectations and allow managers to reconsider alternative methods when it comes to saving behaviours and adapt their marketing strategies in accordance (Souiden & Rani, 2015), while knowledge and understanding of the Pomaks' characteristics can assist in social interventions aiming to provide guidance in the pursuit of a healthy household's financial life (Satsios & Hadjidakis, 2018). The study scope was limited to the Pomaks in the Greek side of Thrace, but there are Pomak populations in the Bulgaria side as well. Consequently, future research could include comparative studies involving Pomaks on other countries (such as Bulgaria).