Exploring Consumer Reactions to the Recession in Slovenia and Bosnia and Herzegovina
Tomaž Kolar Vesna Žabkar Nenad Brkic Ismir Omeragic
In order to analyze the impact of the current economic crisis on consumers of FMCG products in the Balkan region this paper investigates changes and trends in consumer values and attitudes between years 2007 and 2009 in Slovenia and Bosnia and Herzegovina. A comparison of value ranks and factor scores between the two years and two countries shows some significant differences and surprising findings in terms of changes in consumer values and attitudes, which were otherwise expected to be relatively stable over a short time period. In order to provide an insight into questions of if, why and how the crisis differently affected Slovene and Bosnian consumers, possible causes for identified differences are discussed from the theoretical and practical standpoint and several managerial implications are suggested.
Key Words: recession, consumers, attitudes, values, Slovenia,
Bosnia and Herzegovina jEL Classification: M31
Introduction
In order to adjust strategies to changing marketing conditions, companies were recently advised to analyze how customers' needs and spending patterns shift during and after economic crisis (Quelch and Jocz 2009; Flatters and Willmott 2009). This wisdom is not new. More than a quarter of a century ago Piercy (1983) asserted that recession has affected how consumers behave as well as how much they spend, which is a vital con-
Dr Tomaž Kolar is an Assistant Professor at the Faculty of Economics,
University of Ljubljana, Slovenia.
Dr Vesna Žabkar is a Professor at the Faculty of Economics, University
of Ljubljana, Slovenia.
Dr Nenad Brkic, is an Associate Professor at the School of Economics
and Business, University of Sarajevo, Bosnia and Hercegovina.
Ismir Omeragic is a Consultant at Valicon d. o. o., Slovenia.
Managing Global Transitions 10 (2): 121-144
sideration in applying the right strategy. According to him the companies will have to spend more time, money, and effort on market research during crisis. Yet remains the question of whether companies do follow these suggestions and whether existing knowledge on how consumers behave during recessions is sufficient in order for companies to develop effective strategies to cope with recessions. The literature review unfortunately reveals that the existing body of knowledge on this topic is rather scarce and narrowly focused. It often deals solely with consumer behavior in terms of aggregate analysis of current and future (down) spending (see e.g. Eppright, Arguea and Huth 1998). Such (macro) analyses are, however, not sufficiently precise in order for companies to be able to use them for development of marketing strategies. One cannot expect that consumers will be consistent in their reactions and down-spend all product categories equally, as is the commonly held assumption in macroeco-nomic analyses. Contrary to this assumption Curtin (2003) for instance predicts that the 21st century consumer will be inconsistent in his expectations and spending patterns. Another problematic assumption that needs to be tested is whether consumers are all equally rational, thrifty and sober during recession, or this could be the vital information for strategic decisions like segmentation. Despite some attempts to segment consumers on this basis, this is often done on an ad hoc basis, without sufficient empirical arguments (see e. g. Quelch and Jocz 2009). Comprehensive knowledge which could explain why different consumers react differently to a recession thus still seems fairly limited. Therefore, the purpose of this paper is twofold.
First, in order to probe behind consumer observable reactions, the paper is focused on two general concepts that could explain and predict consumer behaviors, namely on consumer attitudes and values. Consumer attitudes and expectations towards the economy were already found to be powerful predictors of consumption on aggregate level (Ea-saw, Garratt and Hearvi 2005; Eppright, Arguea and Huth 1998). Much less is known about these two concepts in this regard on a micro level; that is, how more specific attitudes and consumer values are affected by recessions. In order to provide better insight into this issue, a comparison of various consumption related attitudes and general values will be done in two time periods: before and during the current economic crisis. The second purpose is related with the assumption that differences among consumer reactions to a recession are dependent upon various factors like economic development and political stability, but also upon
cultural differences. For this purpose we compared the attitudes and values of consumers in two countries that exhibit many similarities and differences, namely Slovenia and Bosnia and Herzegovina (bh). In what follows, a brief theoretical background of the concept of consumer attitudes and values is presented, together with the existing knowledge on how they are related with the consumer reactions to the recession in both countries. In the empirical part a methodology and results of comparative analysis are presented, followed by discussion of findings and managerial implications.
Literature Review
In the domain of macroeconomics, the analysis of consumer attitudes has a long tradition, where they are termed consumer sentiments. Analyses of trends of various consumer sentiment indexes indicate that these aggregate measures affect the aggregate consumer expenditures more often than the economic indicators and might provide predictive power not possessed by the economic indicators (Eppright, Arguea and Huth 1998). These studies found that consumer attitudes correspond with various macroeconomic indicators of economic growth/recession like, gdp, unemployment, inflation rate, etc. (Curtin 2003). Still, caution is needed at (over)generalizations of the attitude-economy relationship. According to Curtin (2003), consumer confidence is a fairly accurate predictor of macroeconomic change, where some correlations with actual expenditures are stronger (e. g. for home) than others (e. g. for vehicles). Similarly Easaw, Garratt and Heravi (2005) found that consumer sentiment does predict the households' consumption of durable goods, where predictions based on the uk indicators are better than predictions based on the us indicators. Based on this treatment of attitudes as a monolithic variable it seems inappropriate, as consumer expectations regarding various economic and personal/financial issues (jobs, incomes, interest rates, inflation, stock process, home values, etc.) exhibit divergent changes and correlations with actual expenditures, but also depend upon national factors.
Examination of consumer attitudes and their relationship with economic prosperity/crisis warrant attention from the conceptual aspect as well. The macroeconomic approach treats consumer attitudes as a concept which could be measured by a single question. The mori 'confidence' measure, for instance, is a straightforward index based on a single question which asks consumers 'Do you think the general economic
condition of the country will improve, stay the same, or get worse over the next 12 months?' (Easaw, Garratt and Heravi 2005). Consumer attitudes in (macro)economy are implicitly understood as a simple, one-dimensional construct that is synonymous with consumer's estimations, perceptions, predictions, expectations, confidence, sentiments, etc. - in regard to the state of economy. While detailed conceptual distinctions between attitudes and these concepts are beyond the scope of this paper, it is obvious that this is a rather big (over)simplification if not misunderstanding of the attitudes concept.
Attitude is defined as the enduring organization of cognitive, but also emotional and motivational processes, with some link to our environment. It is a learned predisposition to respond in a consistently favorable or unfavorable manner with respect to a given object (Hawkins, Best and Coney 1989). For the purpose of explaining how attitudes affect and predict actual behavior towards a certain object, various models have been used (see Schiffman and Kanuk 2004). According to the three-component attitude model, attitudes consist of a cognitive component (knowledge, perceptions and experiences), affective component (emotions and feelings) and conative component (likelihood or tendency of taking a specific action). On the other hand, multiattribute models explain attitudes with regard to specific attributes of the particular object. Among these, the Fishbein multi-attribute model is considered to be the most influential and measures three components of attitude, namely salient beliefs (key attributes of an object that are considered during its evaluation), object-attribute linkages (the perceived probability that the object has a key attribute) and the evaluation of the important attributes. For the purpose of the improved predictive ability of attitudes, Fishbein's model was revised and improved with several moderating variables (attitude to behavior, subjective norm, perceived control, behavioral intention), which resulted in the theory of planned behavior. Regardless of the various conceptualizations and models that could be adopted for the purpose of investigation of attitudes, 'macroeconomic' conceptualization seems too narrow. In our case such oversimplification is problematic at least from two aspects. Firstly, attitudes are complex evaluative constructs, and reduction to a single dimension might importantly hinder its explanatory and predictive value, if we measure only a certain/narrow aspect (e. g. cognitive) of it. Besides, it is of vital importance how attitudes are measured. Formulation of questions (items) that are used for measurement might namely impact the attitude-behavior relationship. In a
well known experiment consumers were for instance asked 'to evaluate the taste of two cola drinks' (Pepsi vs. Coca-Cola) and preferences were clearly on Pepsi's side. However, when respondents were asked 'which cola they would choose to drink,' Coca-Cola was the clear winner (see Gladwell 2002). This suggests that the aspects of attitudes that are evaluated might be decisive for the outcomes and consequences of evaluation. Consequently the more complete, comprehensive and holistic is the conceptualization and measurement of the attitudes employed, the more firm, valid and reliable are the established relations of attitudes with various consumer behaviors - reactions to crisis in our case. Secondly, attitudes might be focused on various objects and issues. While macroeconomics is mainly interested in consumer attitudes toward 'general economic conditions,' companies are probably more interested in how recession affected consumer attitudes toward certain product categories, particular brand(s), consumption pattern(s), preferred product/brand attributes, choice criteria, etc. Marketers might also be interested in how crisis affects attitudes towards marketing, advertising, pricing, and so on. Collectively these issues are termed consumerist attitudes and pertain to various positive and negative evaluations of marketing issues and activities. While lack of research that would investigate the impact of recession on consumerist attitudes is evident, the available literature indicates that consumerist attitudes are related with economic development and levels of inflation (Lysonski and Durvasula 2003).
Next to attitudes, values are another relevant concept for investigation of how consumer reaction to a crisis might be explained. Values namely share some similarities, but also some important differences in comparison to attitudes, which makes them a complementary and relevant concept for the purpose of the study. Chow and Amir (2006) point out that within the turbulent global economy and marketplace, consumer values tend to be more stable and innate than other human characteristics. Values are considered to be an important evaluative and adaptation mechanism, which impacts the attitudes people hold and is antecedently related to overall attitudes (Wharton and Harmatz 1995). In a study which examined consumer responses to the economy these authors found important differences among consumers for whom inner directed values are more important than for consumers with other-directed values. They found that internal oriented respondents are more information sensitive and are more inclined to use their perceptions of economic indicators (interest rates, inflation, etc.) as factors in their own economic
behavior. According to Chow and Amir (2006) motivation for more specific decisions like brand selection is also deeply rooted in our guiding principles, thus in our values. Values are defined as beliefs about some desirable end-state that transcends specific situations and guides consumer behavior (Salomon et al., 2002). For the conceptualization and measurement of values, several models and corresponding instruments have also been developed, including the List of Values (lov), Values and Lifestyles (vals), the Rokeach Value Survey and the Schwartz values survey (Schiffman and Kanuk 2004). Rokeach (1973) identified two basic and complementary sets of values which are a set of terminal and a set of instrumental values. Terminal values are desired end-states or personal goals that apply to various degrees in many different countries, while instrumental values represent the basic approaches or actions needed to achieve terminal values. A different approach to finding (cultural) value differences was taken by Schwartz et al. (2001). Schwartz did not ask for preferred outcomes (end-states), but asked respondents to assess 57 values as to how important they felt these values were as 'guiding principles of one's life.' Schwartz's typology consists of 10 distinct value types (power, achievement, hedonism, stimulation, self-direction, universal-ism, benevolence, tradition, conformity and security) in an individual-level analysis. Rokeach's and Schwartz's value surveys have been extensively used in the marketing domain (Salomon, Bamossy and Askegaard 2002). Pitts and Woodside (1983), for instance, found that Rokeach's value structures were found to be linked to choice criteria, beliefs, intentions and past behavior for various products and brands. On the global scale Chow and Amir (2006) found six segments according to the value structure (strivers, fun-seekers, creative, devout and intimate) that can be further condensed to two underlying dimensions of hedonism versus pro-social and conformity versus self-direction. The distribution (size) of particular segments differs considerably between various countries. In countries like the uk and Canada, fun-seeker and intimate are often dominant, while in a cluster of Eastern European countries (Czech Republic, Poland, Hungary and Russia) the strivers and intimate segments are dominant. Various national, economic and cultural factors thus impact the prevailing value structure and consequently we might expect that values and attitudes that explain reactions to a recession are also dependent upon these factors. Values in addition offer interesting conceptual implications for an explanation of reactions to crisis. If the assumption of stability of value holds, consumers' values should be more
stable in recession and change less than corresponding attitudes. In addition, consumers with different values (i. e. internal vs. external oriented) should react differently to a recession.
Empirical Evidences Regarding Consumer Reactions
to a Recession
While there are strong theoretical arguments that relate consumer attitudes and values with recession, there is a lack of empirical investigations of these relationships on a micro-level. Existing evidences often concur with theoretical or commonsense assumptions regarding general trends and the impact of recession on consumers, while there is much less agreement regarding specific reactions, trends and regional/national commonalities/differences. In general consumers react to a recession by spending less (Eppright, Arguea and Huth 1998; Easaw, Garratt and Her-avi 2005; Curtin 2003) and by being thriftier and sober (Flatters and Willmott, 2009). Quelch and Jocz (2009) however suggest that marketers should segment customers according to their recession psychology (from fearful to carefree) and how they categorize their purchases (from essentials to expendables). While some segments are expected to be characterized by lower risk of down-spending and extent of the behavioral change during recession, others might be on the opposite side of these two dimensions. Regarding specific products these authors predict that some will not be affected by recession as they are an essential or justifiable expense (e. g. big screen tv), some are expendable and thus their consumption will be reduced (e.g. upscale magazines, health clubs), while sale of some products might in fact increase due to the consumer's concern for safety (e. g. home security) or thriftiness (e. g. frozen food will replace meals in restaurants). Based on more thorough research, Flatters and Willmott (2009) suggest that the following trends are being accelerated during recession: demand for simplicity, higher expectation for ethical business governance, a desire to economize and decreased loyalty toward products and brands. On the other hand four other trends are slowed: green consumerism, ethical consumption, decline in respect to authorities/institutions and extreme experience seeking. Findings of Ang, Leong and Kotler (2000) indicate that consumer countries hit by the Asian crisis in 1997 exhibit some similar trends, namely more comparative shopping, delaying purchases of expensive items, placing more emphasis on product durability and functionality, switching to lower end and local brands, developing a product life cycle cost perspective, relying
more on informative and less on imagery-based advertisements and buying more often at discount stores. Trends reported by Dong-Hun (2009) for responses of Korean consumers to the current recession in general correspond with these findings, while some trends are new or somewhat surprising. As key trends he reports harmony and stability (as a response to anxiety and risks experienced), increased preference for retro products (as a reminder of happier times), polarized consumption (extreme luxury products affordable to a few, while others aspire toward and imitate these few by looking for affordable mass-market luxury), economical consumption (price comparison and discount stores shopping), value shopping in career, education, wealth and infotainment products.
In the absence of entirely comparative information regarding how Slovene and Bosnian consumers respond to the current recession, available findings are presented. In Slovenia, research into consumer perceptions and reactions to the recession in 2009 was intensified, when recession was at its peak. In March 2009 a study titled 'Marketing Monitor' was started and was repeated in September 2009 (dms 2009). From March to September the proportion of Slovene consumers that 'experienced the impact of recession on daily life' had grown from 52% to 63%, while the percentage of consumers that experienced 'lower buying power' had grown from 27 to 37%. More than half of consumers (51%) do 'spend less in the last months.' This percentage however differs substantially across various product categories. In tourism it is as high as 45%, similar (41%) is for sweets, durable products (furniture), clothes and sport (40%) and electronic entertainment products. This is followed by culture and home appliances (34%), cosmetics (31%), mobile telephony (28%), transport (26%), food (24%), education (23%), health (14%) and insurance (11%). In terms of shopping habits Slovene consumers are more prudent and planning, price conscious, they shop more in discount stores and instead of bulk buys they prefer several small buys. What is however surprising is that the percentage of consumers that prefer store brands is decreasing as the recession progresses! Additional (qualitative) research on how consumers perceive recession, performed by Valicon Company enabled additional insights (Zorko 2009). They found that a lot of consumers perceive crisis as being constructed by the media rather than being reality. Recession is often perceived as an external problem, which is not necessarily negative on a personal/individual level (additional benefits offered, lower prices). Still, crisis has a strong psychological effect, as it causes tensions and dissatisfaction. If needed,
the consumers will save on 'less necessary' products, while they are not prepared to downspend on food.
Similar research on reactions to a recession was presented by a GfK company (Mrkun 2009), with some interesting differences. This study briefly addresses consumer values and encompass as the whole cee region, including bh (while results for Slovenia unfortunately were not available). Among the examined countries, recession most strongly affected Serbian consumers - where 61% of them estimate that their financial situation has worsened during the past 12 months (other countries vary from 31% to 60%). In the region product categories that will be mostly affected (rough estimation is that spending will be reduced between 30-50%) by recession are telecommunications, energy, home equipment, transport and restaurants. The category of personal indulgences (coffee, cigarettes, alcohol) will not be substantially affected (2030%), while health, education, cleaning products will be affected least (less than 20%). The GfK report also put forwards the following trends, which are global yet relevant for the investigation of values. They suggest that consumers will try to escape reality as a response to a fast pace of life, and will seek for relaxation after hard work. While some trends are consistent with the previously mentioned (i. e. simplicity, thriftiness, safety), other predicted trends are not entirely consistent (i. e., authenticity, bricolage multicultural ideas and products, youth, individuality, status/success), while some even seem to be counter trends (e. g. ecological consumption, experiences/adventures).
Regarding bh, the GfK study suggests that their consumers are pessimistic as well and are prepared to downspend mostly on expenses for telephone communication, expenses for electricity, water and gas, smoking, tourism and car expenses. Almost 20% of them will also reduce expenses for durable products and household appliances, entertainment (restaurants, theater and cinema). They are however not prepared to give up enjoyment af good food, which will be more often prepared at home.
Other data regarding bh (Omeragic 2009) point out similar conclusions like those in Slovenia (see Zorko, 2009). Recession evokes negative associations like depression, unemployment, loss of jobs and alike. In bh the first things that consumers are prepared to spend less on are 'luxuries' like cigarettes, cosmetics, clothes, tourism. As expected in bh there is a higher share of consumers that personally experience the impact of recession on their daily life (66%) than in Slovenia (52%). Also, recession is in both countries largely perceived as constructed by the mass media
and thus the psychological effect is even larger than the actual effect of recession on personal financial position. This assumption is confirmed by the substantial gap between percentages of consumers that personally experience financial difficulties and the percentages of consumers who actually spend less. The percentage of consumers that already spend less is namely in both countries substantially higher (i. e. 48% in Slovenia and 51% in bh) than the percentage of consumers that personally experience financial difficulties (i.e. 23% in Slovenia and 30% in bh). The current recession thus stimulated a lot of research (and even more speculative predictions) about consumer reactions in terms of behaviors. Despite the fact that the observed reactions indicate possible implications for consumer attitudes and values, they are rather implicit and often inconsistent. More explicit empirical examination of how consumer attitudes and values (might) have actually changed due to the recession is thus needed. There is an evident lack of the studies that would compare consumer attitudes and values before and during recession, but also of the studies that would enable comparison of the differential impact of recession in various countries - what is the focus of the next, empirical part of this paper.
Methodology and Results
In accordance with the purpose of the paper the research design and hypotheses are developed. In order to analyze the effect of the recession, the levels and structure of consumerist attitudes and values of Slovene and Bosnian consumers are compared in the pre-recession period (year 2007) and during recession (2009). In developing the hypotheses, the existing economic conditions are considered (such as political stability and general economic development), but also more specific differences between these two countries that were established previously. In previous studies Žabkar, Kolar and Sunko (2009) and Kolar, Sunko and Žabkar (2009) found important differences between Slovene and Bosnian consumers in terms of consumer attitudes and national character. Bosnian consumers were, for instance, found to hold more positive attitudes towards shopping as fun, peer influence, value for money shopping and on pragmatism in shopping (which denotes lesser emphasise on interior design, differences among brands and store types), suggesting that Bosnian consumers could be described as 'pragmatic hedonists' as opposed to Slovene consumers, who are 'sophisticated rationalists.' In line with theoretical and empirical findings the following hypotheses are proposed.
The first hypothesis is based on the argument that consumer's attitudes and values are mutually related and that both serve as adaptive mechanisms within changing economic conditions (Wharton and Harmatz, 1995; Lysonski and Durvasula 2003; Chow and Amir (2006). Therefore:
H1 The recession significantly affected consumers' values and attitudes, hence the average importance of key values and average consumerist attitudes changed significantly between years 2007 and 2009.
The second and third hypothesis are based on theoretical and empirical evidences, which support the notion that during recession consumers become more thrifty and sober, and that during crisis they suppress the importance of and interest in enjoyment and lavish spending behavior (Flatters and Willmott 2009; Ang, Leong and Kotler 2000; Dong-Hun 2009; ms 2009; Zorko 2009; Mrkun 2009; Omeragic 2009):
H2 In both countries, the value structure is changed in favor of traditional/internal values (as opposed to modern and external values). Hence, it is expected that change in the average importance of traditional/internal values (fairness, safety, health, freedom, self-discipline) will be significantly more positive than modern/external values (materialism, enjoyment, social power).
H3 In both countries, the attitude structure is changed in favor of more economical and cautious consumption instead of hedonistic consumption. Hence, it is expected that the change in average importance for items that are related with economical/cautious attitudes will be significantly higher than for items related with enjoyment/hedonism.
The fourth hypothesis is based on logical argument and available evidence that Slovene consumers, due to their higher incomes, are less affected by recession than Bosnian (Zorko 2009; Omeragic 2009). Hence, the need for adaptation and change of attitudes and values is stronger in Bosnia:
H4 In Slovenia, both the consumerist attitudes and the value structure are less affected (changed), than in Bosnia and Herzegovina. Thus, the average importance of key values and average consumer attitudes changed more in Bosnia than in Slovenia.
Data for analysis were derived from two representative samples of 4612 Bosnian customers and 3219 Slovene consumers in two years (2007 and 2009) by means of a syndicated field survey. The pgm syndicated field survey measures brand-related perceptions, values, shopping attitudes and various other psychographic and demographic characteristics
of consumers. Since the 2 x 2 cross-sectional studies for the two countries and two years shared identical research methodology, including research instruments and field work, they enable comparisons for the examined period and the two countries. Values were measured with 24 items which were taken from Rokeach's (1973) and Schwartz's et al. (2001) value structure. Since our main concern was to identify the most important values for the two countries and how they changed between 2007 and 2009, only selected values that ranked high in importance are reported here (see table 1). All differences in mean values between the two years were significant (p below 0.01) in both samples. It is evident that in both Bosnia and Slovenia values, such as wealth, social power, authority and influence, have significantly increased in importance in 2009 compared to 2007, while, on the other hand, safety of the family, enjoyment of life, concern for the environment, freedom, self-discipline and even health have significantly decreased. Fairness, which was highly valued for Slovenians in 2007, lost importance in 2009 and is now on a comparable level with respondents from Bosnia (for them, it rose significantly between 2007 and 2009).
These findings refer to the second hypothesis, which is not supported. Results not only indicate that during recession traditional/internal values are not becoming more important than modern/external, but the contrary. The majority of traditional values (safety, health, freedom, self-discipline, environment protection) actually became less important during recession, while wealth and social influence became more important. A more detailed interpretation of this finding is discussed in the last part, yet the basic conclusion regarding the second hypothesis seems straightforward. During recession, consumers do not return to traditional values, but rather the opposite - degrade traditional values and embrace modern/external ones (where enjoyment of life is an important exception).
Consumer attitudes to shopping were measured with 20 items covering social influence in shopping (Bearden, Netemeyer and Teel 1990), ethnocentrism (Shimp and Sharma 1987), hedonic shopping (Faber and O'Guinn, 1989), value for money (Lichtenstein, Ridgway and Netemeyer, 1993) and other external influences. Comparison of mean values for consumer attitudes in 2009 compared to 2007 shows some interesting differences (see table 2).
Analysis of the first hypothesis was based on results provided in tables 1 and 2. Given that all important values changed significantly in
table 1 Values: scores for selected items
Items	Year	Bosnia		Slovenia	
		(1)	(2)	(1)	(2)
Fairness (to be frank, natural, genuine)	2007	2.97	2.115	6.01	1.510
	2009	4.14	.840	4-30	.770
Wealth (material possessions, money)	2007	3.03	2.120	2.50	2.107
	2009	3-25	.966	2.86	•792
Social power (control over others, dominance)	2007	2.16	2.359	1.71	2.236
	2009	2.92	1.142	2.63	.928
The safety of the family (security of those we love)	2007	5.77	1.705	5.77	1.745
	2009	4-37	.815	4.28	•839
Authority (the right to lead the other)	2007	2.92	2.416	1.90	2.301
	2009	3-25	1.112	2-59	.909
Enjoyment of life (enjoyment ofleisure, food,	2007	4.12	1.909	4.62	2.050
beverages)	2009	3-64	.918	3-75	•935
Protecting the environment (nature conservation)	2007	4.12	1.834	4.65	1.824
	2009	3.62	.861	3-67	.809
Freedom (freedom of thought and action)	2007	4.94	1.756	5.32	1.758
	2009	3-94	.851	3-99	.862
Self-discipline (sustainability of the temptations)	2007	4.27	1.832	4.19	1.930
	2009	3.66	.874	3-51	.832
Influence (the impact on people or events)	2007	2.87	2.397	2.14	2.283
	2009	3.20	1.103	2.64	.901
Health (to be healthy both mentally and physically)	2007	6.07	1.544	6.10	1.529
	2009	4.60	.699	4.58	.718
Column headings are as follows: (1) mean, (2) standard deviation. Significance for all items (comparing 2007 and 2009) below 0.01.
both countries, and that average values for more than half of all attitude items also changed significantly, these results provide strong support for the first hypothesis. Based on this finding we may assert that both consumers' values and attitudes during recession serve as an adaptation mechanism.
The results in tables 1 and 2 are also relevant for testing of the fourth hypothesis. Regarding attitude change, the results indicate less significant changes in Bosnia (10 out of 19 items) than in Slovenia (11 out
table 2 Consumer attitudes: scores
Items	Year	Bosnia		Slovenia	
		(1)	(2)	(1)	(2)
Shopping is fun	2007	3.28	1.067	2.95	1.116
	2009	3-25	1.018	2.89	1.121
When I go shopping, I cannot stop ***/*	2007	2.60	1.077	2.42	1.092
	2009	2-75	1.011	2-34	1.051
I am interested in what my friends think about	2007	2.56	1.085	2.46	1.056
products I bought ***/**	2009	2.81	1.015	2-59	1.065
I prefer to endorse Bosnian/Slovenian products,	2007	2.99	1.085	3.12	1.077
although this costs me more in the long run -/**	2009	2-95	•983	3.00	1.078
I usually buy in shops where prices are most	2007	3.73	.897	3.56	.986
favorable ***/-	2009	3-57	•927	3-55	.905
When shopping I always look to get as much	2007	3.78	.938	3.72	.939
as possible for my money ***/*	2009	3.68	.920	3-79	.879
Bosnians/Slovenians should buy Bosnian/Slovenian	2007	3.34	1.169	3.21	1.104
products and services	2009	3-38	1.046	3.20	1.117
I like shopping -/***	2007	3.39	1.016	3.18	1.095
	2009	3-34	.970	3-03	1.106
Before purchases I sometimes like to consult	2007	2.87	1.083	2.65	1.086
my relatives ***/***	2009	3-19	•938	3-17	1.005
I often buy products only because they are	2007	2.77 1.089		2.57	1.093
on special offer ***/***	2009	3-07	1.023	2.72	1.089
Continued on the next page
of 19 items). Regarding values, we found that all examined values in both countries changed significantly. More detailed calculation of mean value differences for each country however reveals that the average value change for all 11 values in Slovenia is larger (i. e. 1.00) than in Bosnia (0.75). Based on that we can conclude that the fourth hypothesis is not supported. During recession, the values and attitudes of Bosnian consumers therefore did not change more than the values and attitudes of Slovene consumers.
In 2009, consumers in both countries are significantly more interested in what their friends think about products they buy, also, they like to consult relatives before shopping and often talk to friends about shop-
table 2 Continued from the previous page
Items	Year	Bosnia		Slovenia	
		(1)	(2)	(1)	(2)
I always check prices before I buy a product **/-	2007	3.68	.952	3.59	1.040
	2009	3-59	.926	3-57	.990
I like to watch advertisements	2007	3.11	1.042	2.81	1.180
	2009	3.12	1-039	2.81	1.202
I often talk to friends about shopping ***/***	2007	2.88	1.062	2.48	1.032
	2009	3.01	1.012	2.66	•994
I often buy products only because of the attractive	2007	2.53	1.038	2.22	1.013
packaging ***/**	2009	2.68	.969	2.32	.986
I prefer products of Bosnian/Slovenian producers.	2007	3.04	1.118	2.92	1.070
	2009	3-03	•995	2.91	1.063
When shopping I prefer to consult than to decide	2007	2.77	.996	2.62	.975
by myself ***/***	2009	2.92	•915	2.83	•975
I trust things recommended by experts		3.12	.927	3.06	.945
	2009	3-09	.867	3-05	.888
I am ready to pay more for environmentally	2007	3.18	.977	3.31	.979
friendly products _/*	2009	3-14	.897	3-23	.940
Brands of products that people use tell a lot	2007	3.21	1.010	3.22	.974
about them	2009	3.22	.899	3-17	.987
Column headings are as follows: (1) mean, (2) standard deviation. Significant differences for 2007 and 2009 (Bosnian/Slovenian sample): *** below 0.001, ** below 0.01, * below 0.05.
ping. Both Slovenians and Bosnians would buy products more often because of the attractive packaging. Preference for domestic products has not changed significantly for Bosnians while it decreased for Slovenians. While Bosnians agree with the statement that they cannot stop shopping significantly more, Slovenians agree significantly less. Buying in stores where prices are favorable and checking prices before purchasing has not significantly changed for Slovenians, while it has significantly decreased for Bosnians in 2009 compared to 2007. Slovenians agree significantly more with the statement that they search for value ('to get as much as possible for their money'), while Bosnians agree significantly less. Finally, enjoying in shopping ('shopping is fun') and readiness to pay more for environmentally friendly products has not changed significantly for
Bosnians, while it has decreased significantly for Slovenians in 2009 compared to 2007.
Factor analysis was used for data reduction and summarization of consumer attitudes to shopping for both Bosnian and Slovenian respondents, since our concern was to identify the underlying dimensions of items included in the survey. The analysis was common factor analysis with principal axis factoring with spss 17.0. The factors were rotated (oblimin rotation and Kaiser Normalization due to high correlations among factors). The Kaiser-Meyer-Olkin Measure of Sampling Adequacy was above 0.80 in both Bosnian and Slovenian samples and therefore acceptable. Variables that correlated highly with more than one factor are excluded from further analysis; leaving 19 variables in the analysis for the Bosnian respondents and 17 variables for the Slovenian respondents. The number of factors and variables was selected based on eigenvalues exceeding 1 and factor loadings exceeding 0.6. Factor loadings for the two samples are presented in table 3.
The obtained factors in both samples can be interpreted as follows: 1. Social influence (explains 24.3% of the variance for the Bosnian sample and 22.8% for the Slovenian sample); 2. Ethnocentrism (8.5% of the variance for Bosnia and 14.4% for Slovenian); 3. Value shopping (7.2% and 11.3% of the variance respectively); 4. Hedonism in shopping (4.6% and 8.2%); and 5. External influences (3.3, and 6.6% of the variance respectively). In all, the five factors explain 47.9% of the total 'attitudinal' variance for the Bosnian respondents and 63.5% for the Slovenian respondents. In both samples, the Social influence factor contains items like 'Before purchasing I sometimes like to consult my relatives' and 'When I shop I prefere to consult than make decisions on my own.' The Bosnian sample also contains 'I like to watch advertisements,' which was not significantly related to any factor in the Slovenian sample. In both samples, ethnocentrism includes the following items consistently: 'Bosnians/Slovenians should buy Bosnian/Slovenian products and services,' 'I prefer products of Bosnian/Slovenian producers,' and 'I prefer to endorse Bosnian/Slovenian products, although this costs me more in the long run.' Similarly, Value shopping contained the following items in both samples: 'I usually buy in shops where prices are most favorable,' 'When shopping I always look to get as much as possible for my money,' and 'I always check prices before I buy a product.' Hedonism in shopping includes 'Shopping is fun,' 'When I go shopping, I cannot stop,' and 'I like shopping.' Finally, External influences contain items like 'I trust
table 3 Factor loadings for consumer attitudes for the two countries
Attitudes	Bosnia	Slovenia
e
?
3 a-*
on S
f3	f4	f5	fl	f2	f3	f4	f5
I often buy products only because they are on promotion	0.546	0.629
I like to watch advertisements	0.348	
I often talk to friends about shopping	0.528	0.599
I often buy products only because of the attractive packaging	0.504	0.646
When shopping I prefer to consult than to decide by myself	0.573	0.713
I am interested in what my friends think about products I bought	0.579	
Before purchasing I sometimes like to consult my relatives	0.505	0.635
O	Rnsnians/Slnvftiians shnnlH hiiv Rn sni an/Slovenian nrnrlnrts and servir**«	n	n Rçy	O
I prefer to endorse Bos./Slov. prod., although this costs me more in the long run	0.743			0.847			
Bosnians/Slovenians should buy Bosnian/Slovenian products and services	0.695			0.857			
I prefer products of Bosnian/Slovenian producers	0.856			0.865			
I always check prices before I buy a product		0.575			0.677		
I usually buy in shops where prices are most favorable		0.746			0.817		
When shopping I always look to get as much as possible for my money		0.769			0.853		
Shopping is fun			-0.89			0.908	
When I go shopping, I cannot stop			-0.536			0.748	
I like shopping			-0.735			0.87	
I trust things recommended by experts I am ready to pay more for environmentally friendly products Brands of products that people use tell a lot about them.			0.548 0.656 0.53				-0.709 -0.696 -0.647
Notes: fi - social influence, f2 - ethnocentrism, f3 - value shopping, f4 - hedonism in shopping, f5 - external influences.
I1
>3
TO
a
o
a >3
table 4 Differences in the factor scores for the two years in the two samples
Items	Year	Bosnia		Slovenia	
		(1)	(2)	(1)	(2)
Social influence ***	2007	-0,144	0,929	-0,119	1,007
	2009	0,116	0,840	0,234	0,943
Ethnocentrism	2007	0,000	0,999	0,019	1,001
	2009	0,000	0,852	-0,038	0,997
Value shopping***	2007	0,060	0,874	-0,023	1,013
	2009	-0,048	0,891	0,045	0,972
Hedonism in shopping	2007	0,004	0,955	0,042	1,004
	2009	-0,004	0,902	-0,084	0,988
External influences	2007	-0,004	0,871	-0,006	1,010
	2009	0,003	0,792	0,012	0,980
Column headings are as follows: (1) mean, (2) standard deviation. Significance: *** below 0.001, ** below 0.01.
things recommended by experts' and 'Brands of products that people use tell a lot about them.' We need to point out that Hedonism is a negatively oriented factor in the Bosnian sample, while External influences are a negatively oriented factor in the Slovenian sample, therefore high values on either of the dimensions a mean low level of agreement with statements that compose the two factors.
The test of differences in the factor scores for the two samples revealed some interesting results. For Bosnians, the social influence has significantly increased in 2009 compared to 2007, while value shopping significantly decreased. Differences for other factors (ethnocentrism, hedonism, external influences) were not significant for the two years in the study. This means that Bosnians in 2009 significantly more appreciate the opinion of friends and relatives compared to 2007, while they do not claim that they would buy more in stores where prices are lower or that they would always check prices before buying a product. Slovenians, similarly, significantly more appreciate the opinion of friends and relatives compared to 2007, on the other hand, shopping is less fun in 2009 compared to 2007. There were no significant differences in other factors examined (see table 4).
Such detailed examination of changes in consumer attitudes is relevant for the dis/confirmation of the third hypothesis. Here, however, the obtained results are not consistent and hence not conclusive. From ta-
ble 2 we can concludes, that some aspects of attitudes changed as expected (Slovene consumers, for instance, enjoy less in shopping during recession), while some changes are in the opposite direction as predicted (Bosnian consumers for instance, enjoy more in shopping during recession). Perhaps the biggest surprise is finding that Bosnian consumers are less price-sensitive in recession (!). An attempt to interpret this finding is provided in discussion, while in regard to the third hypothesis we might conclude that it is not supported. The obtained results therefore do not support the conclusion that consumers during recession develop more rational and sober attitudes.
Discussion and Implications
The presented results provide an interesting insight into psychological reactions on the current recession by Slovene and Bosnian consumers. Some important and 'taken for granted' assumptions are supported, while some of them are challenged and require additional discussion. In general, and contrary to assumptions, values seem to be less stable than attitudes, as the importance of most of the examined values changed significantly between 2007 and 2009 - in both countries. Contrary to expectations and arguments in favor of values (see e. g. Chow and Amir 2006), attitudes might be more appropriate segmentation variables when companies aim for long-term segmentation as the basis for strategic decisions. Caution is however needed in such decisions, as there are some significant differences among various attitudes - where attitudes towards peer influence seem to be more affected by recession (i. e. became more important) than other consumer attitudes. An obvious implication here is that companies in recession should rely more on word-of-mouth tools for promotion. As expected, consumers in Slovenia (but not in bh!) experience less fun in shopping. Surprisingly, value for money shopping is not affected by recession in either country. The somewhat speculative explanation and corresponding implication for this finding is that consumers predominantly save money by reduced purchases of selected products and not by purchasing cheaper products and brands (!). Another assumption that consumers in recession tend to buy more local brands (Flatters and Wilmot 2009) also does not seem to be supported, as ethnocentrism is not affected by recession.
Regarding values, the results indicate that the majority of them are significantly affected by recession - in both countries. Here it must be first emphasized that the basic structure of values in both countries remained
approximately the same before and during the crisis. The three most important values before and during recession (in Slovenia and Bosnia) are namely health, safety and fairness. However, the importance of the impact of recession on values was stronger than on attitudes, and their relative importance changed as well. As the importance of most values declined during crisis it does not seems that values play the function of a coping mechanism during crisis and assure psychological stability, but rather that recession further multiplies the erosion of values! Decreased importance is however not equal for all values, but was importantly and differentially affected by recession. As expected, enjoyment is less important in recession, while wealth and social influence - contrary to expectations - remains stable. What is indeed surprising is the finding that traditional values like safety, health, freedom, wisdom and self-discipline are significantly less important in recession. Interestingly, the importance of fairness decreased in Slovenia but increased in Bosnia, which is the only value that actually became more important during recession.
In general consumers thus do not retain and/or return to (traditional, intrinsic) values as expected (and degrade materialistic and status oriented values), but rather decrease their importance in order to avoid psychological imbalance or cognitive dissonance. The results namely show, that while they value enjoyment less, the importance of status oriented values (power, wealth, influence, authority) is not decreased, while traditional values are less important. Rather than social status and other instrumental values, it seems that consumers would prefere to sacrifice, or at least reevaluate the terminal values like safety, health, freedom and wisdom. Putting it another way, it is likely that consumers do not adapt their consumption to existing values, but rather adjust and reframe the structure of values. This finding might have important managerial implications, as value based segmentations might be importantly affected by economical conditions. The size and structure of segments in recession might be more profoundly affected than is commonly assumed. Also, the assumed emphasis on positioning statements might be affected contrary to expectations. The status appeal therefore seems to be more stable than expected, while return to traditional values in appeals might even have adverse effects in promotional efforts.
Given that straightforward interpretations of the results and corresponding implications provided so far are to a large extent surprising, they warrant some additional discussion in order to provide possible explanations. Although such deliberations often open up additional ques-
tions rather than clarify all dilemmas, we are including them as well, as stimulation for future work in theory and empirical research in this field. One such point that requires additional deliberation is the general assumption that crisis affects consumers in such a way that they became more rational and therefore less emotional. Supposedly, and according to available empirical evidences, consumers indeed are thriftier, more sober and spare less for hedonistic, luxurious, impulsive purposes. The question is, however, whether consumers entirely discard or just temporally suppress or perhaps redirect their emotions - and how these reactions affect consumption. From the psychological standpoint it is illogical that the emotional part of consumption would completely vanish. Consumers in fact during crisis exhibit a plethora of different emotions like surprise, concern, fear, anxiety, pessimism, which are all an emotional part of reaction to a crisis. We may assume that these negative emotions could be indeed essential for explaining the often observed irrational reactions of consumers to a crisis. It was for instance found that even if consumers personally do not experience that their economic conditions are worse, they react with downspending (Zorko 2009). Is this just a precaution or fear that causes such reactions? On the other hand, we might expect that consumers would have a need for stability, consolation, for hope and would seek for some positive emotional gratifications, which in the end is reflected in the consumption. How do consumers cope with emotional burdens, where do they find consolation, optimism, how do they compensate for negative emotions? Is it possible that products that satisfy these needs are in fact consumed more during crisis? Quelch and Jocz (2009) suggest that consumers will react rationally and purchase more safety-related products or replace some products with cheaper alternatives (e. g. frozen food will replace meals in restaurants). On the other hand, we can expect that consumers next to (or even instead of!) long-term oriented, prudent and carefully planned purchases seek exit out of crisis in relatively cheap short-term gratifications like alcohol, gambling, smoking and alike? This explanation seems especially relevant when consumers degrade the importance of traditional values as in our case. Again, some products - and services - related with such motives might potentially be more sought in crisis.
Another point which warrants discussion is the point of reference, from which consumers judge the crisis and its consequences. Logically, consumers with lower incomes are normally more adversely affected by economic recession. As in general, incomes in Bosnia and Herzegovina
are lower than in Slovenia, so we might expect that consumers in Bosnia are more strongly affected by recession. The obtained results in general support this notion, but the expected differences are rather small and the trends are very similar. On the other hand, some observations and opinions suggest that the effect of recession on Bosnian consumers is less significant, because they are more familiar with and adapted to crisis, as they are in a state of constant recession, so the current recession is nothing new. As put by some market researchers, their consumers are 'well trained to a crisis, as they have recession every day and all the time' (Omeragic 2009). On the other hand, besides the different economic point of reference Bosnian consumers might judge crisis from a different cultural point of reference. As they live in a different cultural context to that of Slovene consumers, their perceptions and reactions might be framed by their cultural context and hence different for this reason. The main differences were expected in value structure, as Bosnian consumers are considered to be more carefree, relaxed, hedonistic, social and extro-versive than Slovene consumers. However, the value structure is relatively similar in Slovenia and Bosnia, which is yet another possible explanation for the similar reactions of consumers in both countries to a crisis.
References
Ang, S. H., S. M. Leong, and P. Kotler. 2000. 'The Asian Apocalypse: Crisis Marketing for Consumers and Businesses.' Long Range Planning 33 (1):
97-119.
Bearden, W. O., R. G. Netemeyer, and J. E. Teel. 1990. 'Further Validation of the Consumer Susceptibility to Interpersonal Influence Scale.' Advances in Consumer Research 17 (1): 770-6.
Chow, S., and A. Amir. 2006. 'The Universality of Values: Implications for Global Advertising Strategy.' Journal of Advertising Research 46 (3): 301-14.
Curtin, R. 2003. 'What Recession? What Recovery? The Arrival of the 21st Century Consumer.' Business Economics 38 (2): 25-32.
dms. 2009. 'Trženjski monitor dms: jesen 2009.' Društvo za marketing Slovenije. http://www.dmslo.si/media/trzenjski.monitor.dms.-.predstavitev.ms.dms.nov.09.pdf
Dong-Hun, L. 2009. 'Consumption Trends for 2009.' seri Quarterly 2 (1): 100-5.
Easaw, J. Z., D. Garratt, and S. M. Heravi. 2005. 'Does Consumer Sentiment Accurately Forecast uk Household Consumption? Are There Any Comparisons To Be Made With the us?' Journal of Macroeconomics 27 (3): 517-32.
Eppright, D. R., N. M. Arguea, and W. L. Huth. 1998. 'Aggregate Consumer Expectation Indexes as Indicators of Future Consumer Expenditures.' Journal of Economic Psychology 19 (2): 215-35.
Faber, R. J., and T. C. O'Guinn. 1989. 'Classifying Compulsive Consumers: Advances in the Development of a Diagnostic Tool.' Advances in Consumer Research 16 (1): 738-44.
Flatters, P., and M. Willmott. 2009. 'Understanding The Post-Recession Consumer.' Harvard Business Review 87 (7-8): 106-112.
Gladwell, M. 2002. The Tipping Point: How Little Things Can Make a Big Difference. Boston: Little, Brown and Co.
Hawkins, D. I., R. J. Best, and K. A. Coney. 1989. Consumer Behavior: Implications for Marketing Strategy. Homewood, il: Irwin.
Kolar, T., R. Sunko, and V. Žabkar. 2009. 'Exploring the Character and Consumerist Attitudes of the Balkan Consumer.' In Rethinking Marketing in a Global Economy: Proceedings of the 34th Annual Macromarketing Conference, edited by T. H. Witkowski, 7-11. Kristiansand: The Macromarketing Society and the University of Agder.
Lichtenstein, D. R., N. M. Ridgway, and R. G. Netemeyer. 1993. 'Price Perceptions and Consumer Shopping Behavior: A Field Study.' Journal of Marketing Research 30 (2): 234-45.
Lysonski, S., and S. Durvasula. 2003. 'Should Marketing Managers be Concerned About Attitudes Towards Marketing and Consumerism in New Zealand? A Longitudinal View.' European Journal of Marketing 37 (34): 385-406.
Mrkun, M. 2009. 'Obnašanje potrošnikov v casu recesije.' cfk Slovenija, Marketinški fokus, 22 April. http://www.dmslo.si/media/obnasanje .potrosnikov.v.casu.recesije.gfk.pdf
Omeragic, 1.2009. 'Što se zapravo događa?' Presentation at Valicon Knowledge Club, Sarajevo, April.
Piercy, N. 1983. 'What Marketing Lessons Can Retailers Learn from the Recession?' Retail and Distribution Management 11 (3): 35-38.
Pitts, R. E., and A. G. Woodside. 1983. 'Personal Value Influences on Consumer Product Class and Brand Preferences.' The Journal ofSocial Psychology 119 (1): 37-53.
Quelch, J. A., and K. E. Jocz. 2009. 'How to Market in a Downturn.' Harvard Business Review 87:52-62.
Rokeach, M. 1973. The Nature of Human Values. New York: Free Press.
Salomon, M., G. Bamossy, and S. Askegaard. 2002. Consumer Behaviour: A European Perspective. Harlow: Pearson Education.
Schiffman, L. L., and L. L. Kanuk. 2004. Consumer Behavior. 8th ed. Upper Saddle River, nj: Pearson.
Schwartz, S. H., G. Melech, A. Lehmann, S. Burgess, M. Harris, and V.
Owens. 2001. 'Extending the Cross-Cultural Validity of the Theory of Basic Human Values with a Different Method of Measurement.' Journal of Cross-Cultural Psychology 32 (5): 519-42.
Shimp, T. A., and S. Sharma. 1987. 'Consumer Ethnocentrism: Construction and Validation of the cetscale.' Journal of Marketing Research 24 (3): 280-9.
Zorko, A. 2009. 'Vpogled v secesijo na slovenskem.' Valicon, April. Http:// www.dmslo.si/media/fokus.zorko.print.pdf
Žabkar, V., T. Kolar, and R. Sunko. 2009. 'Consumerist Attitudes in Slovenia, Croatia, Bosnia and Herzegovina and Serbia: Distinctiveness Rather Than Commonalities?' Akademija mm 9 (14): 33-40.
Wharton, Z., and H. R. Harmatz. 1995. 'Responses to the Economy among Two Value Orientation Segments.' Journal of Economic Psychology 16 (2): 205-222