CC BY-NC-ND
181
NEW TEXTS
I
P. Martynenko, Y. Roshchina
Patterns of Alcohol Consumption as a Social Group Indicator in Modern Russian Cities1
MARTYNENKO, Petr Alexandrovich —
Student of Master's Program "Applied Methods of Social Analysis of Markets" of Faculty of Sociology, National Research University Higher School of Economics. Address: 20 Myasnitskaya str., Moscow, 101000, Russian Federation.
Email: petr.martynenko@ gmail.com
This article analyzes the correlation between alcohol consumption patterns in Russian cities and the characteristics of consumers, including their social status. The empirical dataset used in this study was generated from the Russian Target Group Indexfor 2000-2010 andproduced by Synovate Comcon. The methods used in the study include correlation analyses, cluster analyses and correspondence analyses. The results of the study confirm that differences in alcohol consumption patterns are important characteristics of social groups — stratified by gender, age, education and income — in Russia. Beer, vodka and other spirits are typically consumed by men, whereas wine, champagne and liquors are typical consumed by women. The different social classes also have different chosen beverages: the highest social classes prefer wine, champagne, cognac, whisky and exotic beverages such as rum and tequila. The volume of consumed alcohol is not an indicator of social class. Beer and vodka — beverages consumed by all social groups — are mostly consumed by the poorer and less educated. This study also identified the following consumer clusters: "light drink lovers" (beer-oriented consumption), the "masculine consumer" (consumption of beer and vodka), the "feminine consumer" (wine- and champagne-oriented consumption), and the "eclectic type" (multi-oriented consumption). These clusters have different social and demographic characteristics. In modern Russia, patterns of alcohol consumption and the social class of the consumer tend to be highly correlated. Variegated consumption patterns associated with the postmodern lifestyle were detected in fewer than 5% consumers of alcohol; these consumers tended to be educated, well-off, young and employed in executive positions.
Keywords: consumption structure; alcohol consumption; social class; social status indicators; lifestyle; postmodern society.
The deterioration of health, the increasing mortality rate and the social consequences (e.g., violence and loss of income) associated with excessive alcohol consumption have prompted researchers and state authorities to place increasing emphasis on the task of reducing alcohol consumption [WHO 2012]. In recent years, Russia has adopted disincentives for alcohol consumption, including increasing minimum retail prices, limiting the sale of alcohol at night, instituting a ban on the sale of alcohol in kiosks, requiring licenses from the manufacturers, etc. [Neufeld, Rehm 2013]. In addition to its numerous negative impacts, alcohol is addictive — in other words, the current consumption level depends on the previous consumption level. However, alcohol is a consumer
The results of the project "Sociological Research of Contemporary Russian Markets", carried out within the framework of the Basic Research Program at the National Research University Higher School of Economics (HSE) in 2014, are presented in this paper.
ROSHCHINA, Yana Mikhailovna —
Candidate of Science in Economics, Associate Professor of Department of Economic Sociology, Senior Research Fellow, Laboratory for Studies in Economic Sociology, National Research University Higher School of Economics. Address: 20 Myasnitskaya str., Moscow, 101000, Russian Federation.
Email: yroshchina@ hse.ru
good and its consumption depends on the economic and social patterns of consumption of other products (such as foodstuffs): many studies have shown that alcohol consumption increases with income and decreases with increasing prices (ref.: [Grossman et al. 1993; Clements, Yang, Zheng 1997; Andrienko, Nemtsov 2005] et al.). Empirical research has also shown that factors such as gender, age, education and place of residence influence the demand for alcohol. (ref.: [Farrell, Fuchs 1982; Hughes et al. 1997] et al.).
In addition to studying quantitative aspects of the demand for alcohol (which, as a rule, are associated with levels of ethanol consumption), it is necessary to analyze patterns of consumption (i. e., the percentage of consumers consuming certain drinks and the volumes consumed by these groups). In the 1990s, Russia shifted from a predominantly so-called northern pattern of consumption (primarily vodka-oriented) to an average European pattern of consumption (beer-and vodka-oriented) [Roshchina 2012: 245]. By the early 2000s, a radical change in the patterns of consumption had occurred: the amount of vodka and homebrew consumed decreased and the percentage of people consuming beer and wine increased. In the last 10 years, the percentages of consumers consuming different alcoholic beverages have remained relatively stable. This change has often been regarded as positive; however, because beer has a lower-proof potency than vodka and therefore tends to be less fatal, "beer alcoholism" has become a topic of concern [Shatikhin 2012].
The changing consumption patterns are important contributors to the reduction in the general volume of alcohol consumed and, consequently, to the improvement in public health. The choice of alcoholic beverages and other consumer goods (including foodstuffs) reflects the lifestyle and social status of the consumer. For this reason, our study focused on patterns of alcohol consumption and the social characteristics attributed to different beverages. Many studies have stressed the need to differentiate foodstuffs from beverages when evaluating social status indicators [Bourdieu 1984; Douglas 1987; Thornton 1987]. However, in the last quarter of the 20th century, sociologists described the emergence of the "post-class" society where individual practices are less and less defined by the usual social units (class, stratum, gender, age and group) [Beck 1992: 2-3]. Russia is often described as a "post-class" society [Ionin 1998: 252]. This phenomenon is corroborated by the theory of postmodern society where lifestyle becomes ever more fragmented and disconnected from social statuses [Harvey 1989; Herpen, Verger 2008].
In this study, we evaluate whether the consumption of specific alcoholic beverages is a marker of differentiating social groups in modern Russia. First, we will present an overview of the economic and sociological concepts of alcohol consumption that explain which factors influence the volume and patterns of consumption. Second, we will consider the methodology used in this study. Finally, we will present the empirical results.
Theories of Alcohol Consumption in Economics and Sociology
The reasons for, features of and factors associated with the consumption of alcoholic beverages have been the subject of economic and sociological studies
for quite a long time. We will consider the concepts that describe which social groups (in terms of gender, age, occupation, and hierarchy) tend to consume a larger or smaller volume of alcohol and specific beverages and brands.
In economics, empirical research into this field began with the publication of G. Becker and K. Murphy; the authors developed the theory of rational addiction to explain the commitment of an economic agent to a specific good over a long period of time [Becker, Murphy 1988]. Drawing on the notion of consumer capital, the researchers show that a habit is the result of maximizing the benefit of consuming a good. This result is related to the accumulation of consumer capital [Becker, Murphy 1988: 677]. An important conclusion of this model is that people focused on the short term are more likely to consume "harmful" goods than people focused on the long term [Becker, Murphy 1988: 682]. It is therefore likely that individuals with a higher level of education are less likely to consume alcohol because they are better at forecasting their future.
Another aspect of the study of alcohol consumption is the connection between the level of consumption, the income of the consumers and the pricing of the goods. The consumption (i.e., the physical volume) of alcohol increases with income. However, certain researchers have reported a non-linear correlation between income and consumption. A U-shaped correlation between income and consumption has been identified in Russia: the poor and the rich drink more than individuals with an average income [Andrienko, Nemtsov 2005]. Another attribute of alcohol consumption is that the amount spent on alcohol (but not the level consumed) increases with the income of the consumer. This means that more well-off individuals prefer to improve the quality of the consumed spirits and purchase better more expensive beverages instead of increased quantities of alcohol.
In the social bond theory, alcohol consumption is perceived as a form of deviant behavior [Hirschi 1998]; another sociological approach perceives it as a reaction to the unhealthy psychological and emotional condition of an individual [Peirce et al. 1994]. These ideas suggest that the poor, the social dropouts and people suffering from stress at their workplace (for instance, top managers of enterprises) may be more active consumers of alcohol.
In this article we are more interested in the so-called structural approach, which focuses on the type (and not just the volume) of the consumed goods. This approach was developed to analyze the correlation between the social structure and culture of a society and the variety of its gastronomic preferences (including beverages) [Douglas, Isherwood 1979; Levi-Strauss 1981]. Here, the consumption of different kinds of alcohol is tightly connected with the gastronomic preferences of the region, time and social class [Braudel 1979]. N. Elias [Elias 1969, 1982] and S. Mennell [Mennell 1987] have also shown that changes in nutrition (and in the consumption of beverages) are subject to the so-called civilizing effect; according to this effect, consumers exert increasing control over their affects, and eating and drinking becomes more and more restricted and socialized. At the same time, this theory introduces limits to certain kinds of food and beverages, including limits related to the markers of gender and social status [Mennell 1985; 1987]. In this way, the consumption of meat and certain kinds of alcoholic beverages becomes a privilege of the rich or of men. Strong drinks are also a symbol of masculinity. A possible explanation for why women and young people are not allowed to consume strong drinks in many societies is that limiting their consumption protects their health and that of their offspring [Douglas 1987: 7]. In this context, the consumption of alcohol (strong drinks especially) might symbolize maturity for young people or gender equality for women.
In cultural studies and in anthropology, eating and drinking represent a ritual. Alcohol may be consumed when an employee has transitioned from a work setting to a leisure setting; alcohol can also be a marker of social exclusion, of social roles or of friendship [Douglas 1987: 10]. In addition, the consumption of alcohol is a ritual that is often associated with important events. Certain beverages are associated with certain holidays (champagne and the New Year, for example). Thornton examined the consumption of two alcoholic beverages
in Austria — sekt and schnapps — in the context of their social meaning [Thornton 1987]. Sekt (sparkling wine) is a formal drink consumed during the holidays and important celebrations. In contrast, schnapps (fruit vodka) is not associated with formal events and is typically consumed among family and friends. The consumption of specific drinks associated with a certain social relationship can say a great deal about the consumer. For example, sekt is typically consumed by professionals who prefer formal relations and are more individualized, whereas schnapps is more often consumed by workers and farmers who tend to establish stronger more friendly social relationships [Thornton 1987]. In this context, alcohol consumption is no longer a means of satisfying an individual need: it becomes a means of reproducing the structure of social positions and acquires a symbolic meaning [Baudrillard 1968].
Approaching the consumption of goods (and of alcohol in particular) as a mechanism of reproducing the social structure is one of the basic ideas put forth by the French sociologist Bourdieu [Bourdieu 1984]. Bourdieu described the effect of lifestyle (practices and tastes) on the volume and balance of the economic and cultural capitals of different social groups. Bourdieu notes that different conditions of life generate different practices. Professors and teachers, who possess a larger volume of cultural capital than industrialists and businessmen, consume different types of food and beverages and pay different relative amounts for these goods. The consumption of wine and beer is more characteristic of workers in France, whereas the more affluent classes typically drink cognac and champagne. At the same time, the theory of postmodern society states that the modern world erases many of these class differences [Harvey 1989]; in a postmodern society, the city dwellers acquire a more fragmented and variegated lifestyle, with practices adopted from different social segments of society.
Empirical research on alcohol consumption is another source of data contributing to the structural approach. The typologies of alcohol consumption used by the authors can draw from both the quantitative and the qualitative approach. The quantitative approach traditionally identifies at least three groups: the abstainers, the moderate drinkers and the harmful drinkers [Brennan et al. 2009]. The qualitative approach distinguishes the northern consumption pattern (choosing beverages like vodka and occasional binge drinking) from the southern consumption pattern (with predominant and regular (but not abusive) consumption of wine) and the central European consumption pattern (focused on the consumption of beer and, occasionally, spirits) [Popova et al. 2007]. Statistical data show that the central European pattern is being embraced in more and more countries, including countries previously attached to the northern or southern patterns of consumption [Popova et al. 2007]. Russia is not an exception to this trend: in the 1990s, the number of vodka consumers dropped considerably, with a resultant increase in beer consumers [Tapilina 2006; Roshchina 2012].
Consequently, the findings of alcohol-related research have generated two competing hypotheses.
Hypothesis 1: the volume and patterns of consumption of alcohol in Russia remain strongly associated with social group (i.e., gender, age, social class, and other characteristics).
Hypothesis 2: the boundaries defining the consumption patterns of social groups are becoming increasingly blurred (especially in the cities), and more citizens are adopting a postmodern lifestyle characterized by combinations of fragments of different types of consumption. In our case, this hypothesis suggests that there is a weak connection between how alcohol is consumed and the social characteristics of consumers, and those consumers drink combinations of different beverages.
Methodology of the Research
For a number of reasons, we will only focus on alcohol consumption in big cities. First of all, differences in alcohol consumption patterns have been shown to vary with the level of urbanization: in big cities, the
percentage of consumers who drink wine and cognac is 1.5-2 times greater than in rural areas, whereas the percentage of homebrew consumers is 2 times smaller [Klimova 2007: 35; Roshchina 2012]. Secondly, social alcohol consumption patterns are more homogenous in rural areas; therefore, testing the correlation between consumption patterns and social characteristics in these areas is of less interest. Thirdly, conspicuous consumption and postmodern lifestyles are more typical for city dwellers than for rural populations [Herpen, Verger 2008; Kozyreva, Dorofeeva 2008: 73].
This project uses a dataset generated by the RusIndex National Survey2 provided to us by Synovate Comcon3. Synovate Comcon has been collecting these data every year since 1995. RusIndex is a wide-scale survey tracking the consumption of goods and services, media audiences and lifestyles, and social and economic parameters of households and their members. The annual sample of RusIndex includes 14,800 households (28,000 individuals aged 10+). The source population consists of 62.5 million people living in cities with a population of more than 100,000 people.
We will focus on the patterns of alcohol consumption using data from RusIndex from 2000-2010 and on factors influencing consumption of different types of alcohol using data from 2010. The source population of RusIndex (people aged 16+ living in cities with a population of more than 100,000 people) consisted of more than 57 million people in 2010; the sample consisted of almost 29,000 people, 70.5% of whom had consumed at least one alcoholic drink in the previous three months. This corresponds to a source population of alcohol consumers of 40.3 million people (sample: 18,560 people).
Another Russian database often used to analyze alcohol consumption is the Russia Longitudinal Monitoring Survey of HSE (RLMS-HSE)4. However, results obtained using data from the RLMS-HSE and RusIndex databases can differ for a number of reasons. Firstly, information on alcohol consumption in RLMS-HSE pertains to the month before the survey was conducted (in most cases, in October or November), whereas in RusIndex, it pertains to the previous three months. In addition, because the data in RusIndex are collected quarterly, the influence of seasonality is limited. This explains why the reported percentage of consumers of champagne and cognac is considerably higher in RusIndex, as these beverages are often consumed during celebrations. Furthermore, these beverages are more typically consumed by city dwellers; RusIndex specifically targets urban dwellers, whereas RLMS-HSE samples the entire country of Russia. We should also mention that RusIndex does not collect data on the consumption of homebrew and homemade wine. Because these beverages are consumed in rural areas on a much larger scale, RuisIndex estimates of alcohol consumption in cities are not biased significantly.
We will begin our analysis with a study of the percentages of consumers of alcohol in general and of different alcoholic beverages in particular to subsequently evaluate changes in consumption patterns. The respondent who reported having consumed the relevant beverage during the three months prior to the survey is considered to be a consumer of this beverage. Correspondingly, the respondents who reported having consumed at least one alcoholic beverage over the three previous months are considered alcohol consumers.
Next, we will analyze the correlation between the consumption of different beverages and the respondent's gender, age, education, income and social class using correlation analysis and correspondence analysis (CA). In the CA, we will use two groups of variables: the consumption of different alcoholic beverages (the first group) and the socio-demographic characteristics (the second group). For different models of CA, the variables in the second group will differ (gender and age groups, professional and income groups, social classes). In our study,
2 For more information on RusIndex see: URL: http://eng.synovate.ru/default.asp?trID=57
3 URL: http://eng.synovate.ru/
4 See, for example: [Roshchina 2012].
we will always follow the social grade classification5 described in the methodology of the European Society for Opinion and Market Research (ESOMAR) adopted in RusIndex, where A is the top class, B, C1 and C2 are middle classes (the highest-middle, middle-middle and lowest-middle, respectively), D is the lowest class and E is the lowest-lowest class.
We will then use the brand repertoire map of the alcohol market showing the correlation between the chosen drink and the average number of all beverages used by a given consumer.
In the next stage of our analysis, we will identify specific types of consumers depending on the patterns of consumption of their preferred beverages by creating a cluster analysis model based on the totality of respondents who consume alcohol. To classify consumers, we used cluster analysis with the method of ^-averages for factors embedded in the DataFriendWeb package using the dichotomous variable "does a respondent drink this alcoholic beverage?"6. For each cluster, we will analyze the importance of correlations between socio-demographic characteristics of the consumers and the "number of cluster" variable.
Finally, we will review the correlations between the consumer characteristics (gender, age, income, education, social class) and the volume, frequency, types, brands and place of consumption of beer, vodka, wine and cognac using correlation analysis and correspondence analysis. In the correspondence analysis, we will use a set of dummy variables indicating whether a certain brand of each drink is consumed by a respondent for the first group.
Alcohol Consumers in General: Dynamics and Factors
In 2000-2010, the percentage of respondents aged 16+ who had consumed any type of alcohol in the urban population fell from 78% to 70.5%, representing a reduction of more than 2 million people (from 42.3 million to 40.3 million). This reduction was particularly reported among men: this percentage fell by almost 11 percentage points among men and by only 5 percentage points among women.
Most of the factors identified to influence probabilities of alcohol consumption remain the same for the populations of big cities throughout Russia [Roshchina 2013]. All of the factors described below have an important effect on alcohol consumption that was discovered by correlation analysis. Here, the probability of consuming alcoholic beverages increases linearly with increasing education level and income per capita (except in the least affluent group). The association between the probability of alcohol consumption and age is quadratic: the percentage of respondents who consume alcohol grows and then falls with age. Alcohol consumers aged 35-44 account for the largest age group of alcohol consumers. In respondents aged 16-19, women account for a larger percentage of drinkers (51%) than men (46%); however, in the next age interval (20-25), men account for a greater proportion of drinkers than women. In 2000-2010, the consumption of alcohol fell among men and women of all ages except those in the eldest age group. This percentage is very similar to that observed in men aged 65+ (75% in 2010); interestingly, the percentage of women of this age who drink has grown (from 40% to 55%).
As in other studies using RLMS-HSE data [Roshchina 2012]), the percentage of respondents who consume alcohol is largest among people living in cities with a population of more than one million, among married people or couples that are not officially married, among the employed (regardless of the type of work - i.e., intellectual or physical) and in households with no children or with children younger than five-years old. Families with children aged 10-18 appear to have fewer consumers of alcohol because parents do not want to
5
Description of classes and the methodology of their construction, see, for example: URL: http://www.frekans.com.tr/eng_ socio_07.html
Unfortunately, data on consumption volumes were not available for all alcoholic beverages, which probably make it less informative for cluster analysis.
motivate their children to start drinking. In this case, the social bond theory of alcohol consumption is more credible.
The percentage of respondents who consumed alcohol was 80.2% among married men and 67.9% among single men (significant correlation coefficients were observed for married men (+ 0.132) and for single men (- 0.124) relative to the Russian average). For women, these differences are less pronounced: 68.5% and 63.3% of the respondents who reported being alcohol consumers were married and single women, respectively. However, the correlation coefficients show a weak (- 0.043 for single women and + 0.061 for married women) but significant correlation: single women are less likely to consume alcohol than married women.
Of the cities with a population of more than one million inhabitants, St. Petersburg had the smallest percentage of respondents who were abstainers (22%). Surprisingly, more alcohol consumers were found among the Ukrainians and Belarusians (75% and 80%, respectively) than among the Russians (70.6%); the percentage of alcohol consumers was similar among the Tatars and Russians (69.5%). This contradicts the assumption that Islam represents an obstacle to alcohol consumption. The percentage of respondents who were drinkers was highest among the members of the highest and middle-middle social classes (approximately 74%) and lowest among the lowest social class (60%).
Patterns of Consumption by Type of Alcohol
From 2000 to 2010, the popularity rating of different kinds of alcohol did not change; however, the percentage of respondents who consumed beer, vodka, wine and ready-made low alcohol cocktails decreased slightly, whereas the percentage of respondents who consumed other beverages increased (see Fig. 1). In 2010, as was the case a decade ago, the ranking of the percentages of respondents who consumed alcohol (from highest to lowest) were as follows: beer, vodka, wine and champagne, cognac, low alcohol cocktails and vermouth. The percentage of respondents who consume whisky and rum grew (two-fold in the case of whisky) in 2005-2010; however, the number of consumers of these beverages remained relatively low (the percentage of respondents who consumed whisky and rum was as low as 1.5% and 0.7%, respectively, in 2000).
80 %
70 60 50 40 30 20 10
0
Beer
Vodka and spirits Wine
Champagne Cognac
Alcoholic coctails
Vermouth
Whisky
Liquor
Gin
Rum
j years
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Fig. 1. Evolution of the percentage of respondents aged 16+ who consumed different types of alcoholic beverage. Ruslndex, 2000-2010.
Alcohol consumption patterns are known to be very gender-specific: beer and vodka are considered the most popular beverages among men, whereas wine and champagne are considered the preferred beverages of women [Roshchina 2012]. However, the patterns of alcohol consumption of women have changed more than those
of men in the past 11 years. For male consumers, the biggest change consisted of a drop in the percentage of vodka consumers (from 81% to 60%) and a smaller drop in the percentage of beer consumers (from 86% to 78%); the percentage of men who consumed cognac and whisky increased (from 13% to 20% and from 5% to 8%, respectively, in 2005-2010).
Similarly, the percentage of women who consumed vodka and beer dropped dramatically (from 53% to 29% and from 60% to 48%, respectively); the percentage of champagne consumers grew from 45% to 51% and the percentage of wine consumers remained the same (52%). The percentage of female wine consumers peaked in 2004 (60% — more than the percentage of beer consumers). In addition, the percentage of women who consumed cognac, vermouth, liquors, and whisky grew from 10% to 18%, from 8% to 11%, from 5% to 9% and from 3% to 6%, respectively, in 2005-2010.
The changes in consumption patterns in the different age groups were also quite significant. The percentage of respondents who consumed vodka in the 20-35 age group fell the most (from 66% to 34%), whereas the percentage of respondents who consumed wine and beer fell less significantly (from 46% to 36% and from 83% to 76%, respectively). The consumption of other beverages did not change significantly among members of this age group. In contrast, the percentage of respondents who consumed vodka and beer in the 55+ age group did not fall as dramatically, whereas the percentage of respondents who consumed wine, champagne and (especially) cognac grew significantly. However, we should note that during the study period (2000-2010), the respondents aged by 10 years: the members of the 20-35 age group in 2010 were aged 10-25 in 2000. Therefore, the change in consumption patterns might be associated with the change of cohorts rather than by changes of preferences within one generation.
Changes in consumption patterns were similar in the lowest and highest (including the highest-middle) socioeconomic classes: a significant drop in the percentage of members of these classes who consumed vodka was observed (from 65% to 45% in the highest A and highest-middle B classes; from 70% to 47% in the lowest D and lowest-lowest E class); the percentage of respondents in these classes who consumed beer changed, but not as dramatically (from 74% to 59% in the A and B classes; from 76% to 67% in the D class; from 67% to 54% in the E class). The change in the percentage of respondents who consumed wine was not significant in any class, even though this percentage grew from 49% to 52% in the A class and fell from 49% to 44% in the B class; the proportion of respondents who consumed cognac grew in all classes (from 25% to 32% in the highest class and from 6% to 10% in the lowest class); the percentage of respondents who consumed champagne fell moderately in all classes except the two lowest classes where the percentage grew (from 25% to 29% in the E class). However, the percentage of consumers of each of these three beverages remains higher in the highest classes.
Factors Influencing the Predilection to Consume Specific Beverages
Correlation analyses using data from 2010 shows that women are more inclined to consume certain alcoholic beverages than men; 52%, 50%, 11% and 8.7% of women drank wine, champagne, vermouth and liquor, respectively; almost all of these percentages are almost twice as large as those associated with men. In contrast, 1.5 times more men drank vodka (59.8%) and beer (78.4%) than women.
The popularity of beer peaks in the 20-24 age group (76%) then gradually decreases to 38.5% in the 65+ age group. The popularity of vodka peaks in the 55-64 age group (54.4%) and falls to 50.3% in the oldest age group. A change in wine consumption patterns was observed in the different age groups. In 2000, the percentage of respondents who consumed wine was highest among young people aged 20-24 (45.5%) and decreased to 34.6% in people aged 65+. In 2010, the percentage of respondents who consumed wine consumers was lowest (29%) among the young people aged 16-19 and gradually increased to 50.7% among the seniors. The cohort effect was not observed here. Similar trends were observed for cognac and brandy.
Using correspondence analyses for the data on beverage consumption, age and gender, we indentified the horizontal axis as gender and the vertical axis as age in the plot (see Fig. 2). Significant positive correlations showed that wine, champagne, liquor and vermouth are the preferred beverages of women aged 25+, whereas rum is the preferred beverage of women aged 20-24. The preferred beverage of men aged 65+ was cognac, whereas that of men aged 25+, men under 54, and men and women aged 20-34 was vodka, beer and whisky, respectively. Tequila and low alcohol cocktails are the preferred beverages of men and women aged 16-24.
In summary, alcohol consumption patterns vary most by gender and age: lighter and sweeter beverages are typically consumed by women, whereas stronger or bitter beverages are typically consumed by men. In addition, certain beverages appear to be typically associated with certain age groups, such as cognac for senior consumers (primarily men), and low alcohol cocktails, tequila, whisky and rum for young people of both genders. The relative novelty of the latter group of beverages on the Russian market may explain their popularity among young people.
We now discuss the correlation analysis conducted to evaluate the relationship between the consumption of different beverages and the social class of the consumer. As we noted in our literature review, many studies have identified a correlation between the consumption of different beverages and social classes. Bourdieu
Only positively correlated variables are shown. Green connector lines mark positive correlations between variables.
7
identifies two types of capital that are unequally distributed among the different social classes in terms of volume and structure: economic capital (corresponding to income) and cultural capital (corresponding to level of education) [Bourdieu 1984]. We use correlation analysis to determine whether correlations between a person's social class (and therefore his or her economic and cultural capital) and beverage preferences exist.
Alcohol consumers with an incomplete secondary education, a secondary education or a professional secondary education mostly preferred beer (over 65%) and vodka (over 45%). Students and high-school graduates also preferred beer (59.9%); however, the second most popular drink among these groups was wine (46.3%). Alcohol consumers with an academic degree preferred wine (63.3%) and vodka (57.3%). Consequently, the percentage of respondents who drink wine and champagne grows significantly with increasing education, whereas the percentage of respondents who drink beer decreases. People with the highest level of education level also appeared to like experimenting: this group had the most consumers of expensive beverages that are not common in Russia (i.e., rum, gin, liquor and whisky).
As we have already seen, beer is the preferred alcohol of consumers. However, beer consumption patterns vary with income level. Groups with an income per capita below RUB 2,000 or between RUB 5,000-7,000 have very different percentages of beer consumers (70.6% and 57.4%, respectively). Increasing incomes result in greater percentages of champagne, whisky, cognac, brandy, tequila and wine consumers. The correlation analysis also detected a correlation between beverage preferences and social class. In the lowest-lowest class (E), the percentage of champagne (29%) and wine consumers (37.2%) is minimal; these percentages gradually increase to 51.8% and 45.3%, respectively, in the highest class (A). In class E and class A, the percentage of whisky consumers and cognac consumers increases from 1.1% to 11.9% and from 10.3% to 31.6%, respectively. The percentage of respondents in these groups who consume gin, rum and tequila increases; however, this increase is not significant, as these beverages are not popular in Russia. The consumption of vodka changes little from class to class; however, a slight non-linear variation can be observed. The lowest (E) class typically has the largest percentage of consumers of vodka and bitters (47%); this percentage falls to 42% in the middle class and reaches 45% in the highest classes.
The second correspondence analysis was implemented to identify correlations between preferences for alcoholic beverages and income per capita and education level (see Fig. 3; significant positive correlations only are shown). In this model, the horizontal axis represents income and the vertical axis charts the education level of the consumer (top-bottom).
Because the education level, income, position and social class of a respondent are all interconnected, the combination of these variables in a single correspondence analysis is not recommended. For this reason, in the two other models, we used the variables "position" (and "occupation" for the unemployed) and "social class"8. These three models enabled us to identify three groups of beverages.
The first group is composed of vermouth, cognac and wine; respondents with high levels of education tended to consume these beverages. Moreover, cognac tended to be consumed by the members of the highest social structures (A, B and C1) and by high-placed professionals and middle managers. Wine is the beverage of choice of educated pensioners. Whisky, rum, gin and tequila are consumed by people with a high level of income per capita; here, whisky is consumed by members of the three highest classes, rum is consumed by the members of the highest-middle class (B) and tequila is consumed by the members of the middle-middle class (C1). Directors of companies prefer whisky and tequila. Vodka, beer and low alcohol cocktails form a group of beverages consumed by people with low income and education levels. Vodka and beer are typically consumed by the members of the lowest classes (D), by people with a secondary education or professional secondary education and by qualified workers and foremen. The lowest-lowest class (E) prefers vodka over
8 Structure of these CA models is not given here due to word limits in the article.
beer. Students and people with an incomplete secondary education (in most cases, students of technical schools and secondary schools) typically consume low alcohol cocktails.
P Incomplete secondary education
Class E
Fig. 3. Correspondence analysis of alcoholic beverages and education, social class and income per capita. Ruslndex, 20109
The patterns of consumption of different alcoholic beverages confirm our first hypothesis: beverages often act as indicators of a social status and are associated with a certain level of cultural and economic capital and a professional status. We can therefore say that Russian alcohol consumers do not yet resemble members of a postmodern society with a variegated and fragmented lifestyle.
Our data also indicate that the type of work of a person significantly influences their choice of beverage: higher percentages of beer consumers and of vodka and bitters consumers are found among people engaged in physical work than among people engaged in intellectual work (74.4% vs. 60.8% and 52.5% vs. 39.8%, respectively). In contrast, the percentage of respondents who consume champagne, wine and cognac is higher among people engaged in intellectual work (44.5% vs. 29.4%, 47.9% vs. 28.8% and 25.2% against 14.9%, respectively). These results are very consistent with the results obtained by Bourdieu in France and by Thornton in Austria [Bourdieu 1984; Thornton 1987]: these authors found that certain types of alcoholic beverages are associated with status and are tagged with "formal" and "informal" meanings. For example, in Austria and Russia, champagne is perceived as a formal and "segregating" drink, whereas vodka and beer are perceived as informal drinks that are consumed among friends and reinforce social bonds.
Only positively correlated variables are shown. Green connector lines mark positive correlations between variables.
9
Combining the Consumption of Different Beverages
The beverages preferred by greater numbers of consumers tend to be combined less with other alcoholic beverages. For instance, half of the beer consumers (the most popular drink in Russia) also drink vodka, and approximately a third of beer drinkers also drink wine or champagne. Other beverages are less popular. Over 70% of vodka consumers also drink beer; one third of vodka consumers also drink champagne and wine, and one quarter of vodka consumers also drink cognac. Half of wine drinkers also drink beer and champagne, and 38% of wine drinkers also drink vodka. The trend is similar for champagne drinkers. More than half of cognac drinkers also drink wine and champagne; over 60% also drink beer and vodka. Over three quarters of the consumers of almost all other drinks (except liquors and vermouth, for which the percentages listed below are slightly lower) drink beer, and over one half drink vodka, wine, champagne and cognac. Consequently, whisky, gin, rum and tequila are not the beverages of choice of loyal consumers; these drinks are preferred by people with diverse preferences and high incomes who can afford them.
A repertoire map for the consumption of alcoholic beverages is shown in Fig. 4: the vertical axis shows the percentage of consumers and the horizontal axis shows the average number of consumed beverages. There is an evident inverse relationship between these indicators, a situation commonly encountered for a diversity of markets (as a rule, this map is constructed for different brands). It is notable that beer was associated with the highest percentage of consumers (62%) but that beer consumers also had the fewest preferred beverages: indeed, beer drinkers only drank 1.8 other beverages on average (2.8 beverages including beer). Similar trends were observed for vodka, champagne and wine: approximately 40% of respondents tended to drink these drinks and consumed an average of slightly more than 2 other beverages. In contrast, cognac drinkers (approximately 20% of respondents) on average drank three other alcoholic beverages. Other beverages were consumed by no more than 10% of respondents; vermouth and liquor drinkers drank approximately 4.5 other drinks, whisky and tequila drinkers drank approximately 4.5 other drinks and rum and gin drinkers drank approximately 5.5 other drinks.
70
60
50
8 40
£ 30
20
10
^Beer
.Vodka
^Champagne
^Wir
Coctails
Cognac
♦ - -^Vermouth
♦ mo
Liquors
^Whisky ^^ .^Tequila
Rumu
Gin
2.6 2.8 3 3.2 3.4 3.6 3.8 4 4.2 4.4 4.6 4.8 5 5.2 5.4 5.6 5.8 6 6.2 6.4 6.6 6.8 7
Average number of beverages
Fig. 4. Repertoire mapping of the market of alcoholic beverages. Ruslndex, 2010
Types of Alcohol Consumers
As mentioned previously, we used cluster analysis to divide the consumers into different segments according to consumption type. Four clusters were identified.
0
Cluster 1: light drink lovers (37.2% of the total number of alcohol consumers). Most respondents from this cluster consume beer (62.5%), very few consume other spirits (for instance, only 5% drink cognac), and none drink vodka. At the same time, 21% drink champagne and 25.7% drink wine.
The percentage of women in this group was much higher than the percentage of women in the overall sample (58%); however, this group remains relatively gender neutral. The percentage of young people under 35 years of age is significantly higher here (47%) than in other groups. Differences in the education levels of members of this group are slight but significant: in general, fewer people with higher education levels are found in this group (44%). Thirty percent of the members of this group are unemployed; the rest of the members are engaged in either physical or intellectual work (in approximately equal proportions). Members of the lowest E class and of the A and B classes account for 7.5% and 12% of the members of this group.
Cluster 2: masculine consumption type (32.5%). Most of the respondents from this cluster are vodka (99.6%) and beer consumers (71.5%). This cluster was designated "masculine" because these drinks are typically associated with a masculine image. Cognac and champagne is consumed by 17.5% of this group, and wine is consumed by 14%.
As may be expected, men account for almost three quarters of this group; 60% of its representatives are aged 35-64, which is very high relative to the entire sample. More than half of the consumers of spirits are seniors, which is consistent with earlier results: as drinkers age, their consumption tends to shifts towards spirits. More than one half of the members of the cluster (56.9%) have a secondary or professional secondary education; 60% of the employed members (43% of the entire group) are engaged in physical work. Members of the lowest and lowest-lowest classes account for 37.5% of the members of this group; the average income per capita in this group is RUB 14,600 (approximately USD 500).
Cluster 3: feminine consumption type (26%). Wine (88.2%) and champagne consumers (85%) are most numerous in this cluster. The consumption of these beverages, as shown above, is typical for women, and these drinks tend to have a feminine image. Vermouth (21.1%) and cognac (34%) drinkers are also well represented in this group.
Women account for 72% of the members of this group, and a wide spectrum of age groups is represented in this cluster. The age distribution in this group is similar to that of the general Russian population; however, there are significantly fewer people aged 20-35 and significantly more seniors aged 55+. Fifty-eight percent of the members of this group have a higher or incomplete higher education; it is therefore not surprising that 69% of the employed members of this group are engaged in intellectual work (48% of the entire group). Almost half of all members of this group belong to A, B and CI classes (highest, highest-middle and middle-middle classes, respectively), which exceeds the average percentage of the representatives of these classes in Russia in general (36%). However, the average income per capita in this group is not very high (RUB 16,500).
Cluster 4: eclectic consumption type (4.3%). This is the smallest group, but a very interesting group nonetheless: the postmodern consumption style appears to be applicable to this group. The percentages of members of this group who consume all types of alcoholic beverages are high: 88% consume whisky, 78% consume beer, 76% consume rum and 28% consume vermouth and low alcohol cocktails. The percentage of members who consume vodka is lower than in the masculine type group (63%); however, significantly more members of this group than Russians in general drink all types of beverages
This group is not strongly gender specific; however, significantly more men (57%) are found in this group than women. Half of the representatives of this cluster are younger than 35 years old, which is more than in any
other group. This group also has the most members with higher levels of education (64% have an incomplete higher education, higher education or an academic degree), the most members engaged in intellectual work (52%) and the most members with the highest level of income (20% had an income per capita over USD 1,000). Twenty percent of this group belongs to the highest classes (A and B) and 62% belongs to the middle classes (CI and C2). This cluster also has the highest number of managers (top managers account for 5.2% of the group and middle managers account for 17%).
Consequently, the different clusters tend to be gender/age-specific and class-specific. The masculine consumption type is associated with men, seniors and people with lower income and education levels. The feminine consumption type is less age-specific and typical of the middle class, i. e., people with a high education level but not the highest income level. The other two clusters are associated with young consumers: these types are not very gender-specific, but significant differences in social statuses are observed among members of this group. Consumers of light drinks are more often unemployed or students and have lower income levels, whereas the members of the eclectic type cluster are typically well-off young people employed in executive positions. We can confirm our earlier hypothesis that alcohol consumption types in Russia are closely associated with social status, gender and age. Even the eclectic consumption type, which most resembles a postmodern lifestyle, is characteristic of the groups with the highest statuses.
Consumption: Volume, Frequency, Brands and Location
As we have observed, certain alcoholic beverages (cocktails, vermouth, whisky, liquors, gin, rum and tequila) are consumed by less than 10% of alcohol consumers. The consumption of these beverages is therefore a marker of social status. Beer and vodka are consumed by men, women, the young, the old, the rich and the poor, albeit in different proportions. Similar trends are observed for wine, champagne and cognac. It is likely that the volume (or frequency) of consumption and the brands and prices of the consumed alcohol vary with social group. Unfortunately, Ruslndex does not provide information on the costs of alcoholic beverages; we will therefore evaluate differences in other characteristics (based on data from 2010).
Beer
It is not surprising that men are more frequent and more avid consumers of beer than women; indeed, on average men drink beer 7 times per month, whereas women drink beer 3.8 times per month; in addition, men drink 6.4 l of beer, whereas women drink 3.8 l per month. Moreover, men more frequently drink strong beers, whereas women more frequently consume light beers. The consumption volume and the percent of consumers who drink a certain type of beverage have a quadratic relationship with age: the maximum volume, 6.3 l per month, is consumed by people aged 25-34; the minimum volume, 3 l, is consumed by seniors aged 65+. Seniors more frequently choose strong beers. More than 20% of young people under 25 years of age consume beer in bars, clubs, etc; this percentage falls sharply with age, reaching 5% or less for people aged 45+. The largest volumes are consumed by the poor (with incomes per capita under RUB 5,000) and by people with middle income levels (RUB 15,000-30,000). As education levels increase, consumption volume drops. The lowest (D) class has the highest consumption volume (5.7 l per month); the highest (A) class has the lowest consumption volume (5.1 l). The middle classes more often than lower classes consume beer in restaurants, pubs and bars (12-13% vs. 4%), whereas the lowest classes more often than middle classes consume beer outdoors (8-9% vs. 5-6%). Drinking alone is most typical for the members of the highest class (18%) and the lowest-lowest class (24%).
Interestingly enough, Baltika is the most popular beer brand in all classes (11% of the highest class and 14% of the highest-middle class choose Baltika 7; 19.5% of the lowest-lowest class and 17.6% of the lowest class choose Baltika 3). However, the biggest differences among the classes (and among people with different income
and education levels) are found in the preferences for the least popular brands: for instance, Velkopopovicky Kozel is the preferred choice of 8.4% of the members of the highest class but only 2.7% of the members of the lowest-lowest class. No significant differences in the average number of consumed brands were observed among the social classes. Correspondence analysis has shown that the most popular brands of the members of the highest, highest-middle and middle-middle class are Staropramen, Velkopopovicky Kozel and Holsten, respectively; the members of the B and C1 classes prefer Heineken and Carlsberg, respectively; members of the A class prefer Zlatopramen and Sol. In addition to Baltika, members of the E and D classes are also partial to Zhigulevskoe, Klinskoe, Yarpivo, Okhota and Tri Medvedya. These results show that there are two distinct types of brands: the members of the lowest classes prefer cheaper Russian brands, whereas the members of the highest and middle classes prefer more expensive foreign brands. For the members of the lowest classes, it is significantly more important for a brand to be easy to find on the shelves (18.6% of the members of the E class prioritize this feature vs. 8% in the A class); an affordable price is also important (39% of the members of the E class prioritize this feature vs. 23% in the A class); the highest classes care most about the country of origin of the beverage (18% vs. 15% of the E class).
Vodka
As is the case with beer, vodka is associated with a masculine image. It is therefore logical that men consume vodka more frequently than women (on average 3.7 times per month vs. 1.7 times) and consume twice as much as women (an average man consumes 0.6 l per month; an average woman consumes 0.3 l). The frequency of vodka consumption gradually increases with age, reaching 3.3 times per month for seniors aged 55-64, whereas the maximum quantity of vodka (over 0.5 l) is consumed by the people of the middle age group (25-54). The volume of vodka consumed has a non-linear correlation with income and falls with higher levels of education. The lowest classes consume more vodka than the highest classes. Significant differences in the tendency to consume vodka in bars or restaurants can be observed: more young people tend to consume vodka in these places, whereas few members of the lowest classes do. For the lowest classes, an affordable price is a significant priority (this factor is prioritized by 45% of the members of the E class and by 40% of the members of the D class but by only 30% of the members of the highest classes); the ease with which a brand can be found on the shelves is also important (prioritized by 27% of the members of the E class and by 11% of the members of the A class). Interestingly, the most popular brand of vodka in all social classes (except in the lowest-lowest class) is ZelenayaMarka (chosen by 17-18% of the members of the A, B, C1 and D classes); in the E class, the preferred brand is Pshenichnaya. A significantly larger percentage of the members of the highest classes and of people with higher education levels choose Parliament.
Wine
Although the percentage of women who consume wine is higher than that of men, men consume 1.5 times more wine and consume wine more frequently than women (men consume 0.6 l per month and drink 2.7 times per month). Men tend to prefer dry wines, whereas women prefer sweet wines. The consumption volume has a quadratic relationship with age (people aged 35-45 consume the greatest volumes of wine, 0.5 l per month) and a near-linear relationship with income (direct dependence). People with an incomplete secondary education consume more wine (0.6 l) than people with a higher education (0.43 l); however, 16% of the former group and 35% of the latter group consume wine. The members of the highest class consume wine most frequently (2.5 times per month), followed by members of the B and E classes (2.2 times per month). However, the volume consumed in the different classes is not significantly different from that consumed by the average Russian.
Preferences for certain countries of origin differ significantly. The correspondence analysis showed that the members of the highest class prefer German, Chilean and Spanish wines; the members of the highest-middle
class prefer Spanish and French wines; the members of the middle-middle class prefer Chilean, Italian and Argentinean wines, i. e., foreign wines. The members of the lowest-lowest class prefer Russian and Hungarian wines; the members of the lowest class prefer Russian and Bulgarian wines. In general, people with a higher income and education level prefer more expensive foreign wines; Russian wines and cheap wines are preferred by people with a lower income and education level. The members of the highest classes prefer dry wine; the members of the lowest classes prefer fortified wine. As was the case for vodka and beer, the members of the lowest classes care more about the affordability and availability of the wine.
Cognac
Cognac is also considered a more masculine drink: men consume it more frequently (1.8 vs. 1.3 times per month) and consume larger volumes (0.3 l vs. 0.2 l) than women, but the difference is less pronounced than for vodka and beer. The percentage of respondents who consume cognac is highest in the 35-64 age group; however, the largest average volumes are consumed by people aged 20-44 (more than 0.25 l per month). The percentage of respondents who consume cognac grows linearly with the level of social hierarchy (from 6% in the E class to 24% in the A class); however, the individual consumption volume does not change significantly with class. The members of the highest classes prefer more mature cognacs and tend to choose brands for their good taste and quality. The members of the highest classes (A, B and C1) consume Ararat more frequently than other classes (12-13%); the members of the lowest-middle class tend to choose Hennessy (11%), the members of the lowest class prefer Belii Aist and the members of the lowest-lowest class prefer Rossiysky and Belii Aist. The correspondence analysis showed that Remy Martin is also preferred by the members of the highest class.
Conclusions
The results of our study show that alcohol consumption patterns characterize the different social groups of Russia, with different patterns associated with different genders, ages and educational and income groups. This observation allows us to reject Hypothesis 2 and accept Hypothesis 1. The choice of beverages varies by group: beer, vodka and other spirits are typically consumed by men, whereas wine, champagne and liquors are typically consumed by women. In general, women choose lighter and sweeter drinks (and prefer sweet wines to dry wines), whereas men prefer strong and bitter alcoholic beverages. These preferences can be attributed to cultural traditions in which these gender status symbols are rooted. Age differences also reflect the different stereotypes: the senior groups choose vodka and wine; the junior groups choose beer and low alcoholic cocktails. These preferences are most likely the result of aggressive advertising conducted in the late 1990s, affordable prices and convenient dosages: one beer or cocktail (a can or a bottle) is easy to buy and drink, whereas stronger drinks generally require portioning (drinking with other people or over a prolonged period of time).
Social differences related to unequal distributions of cultural and economic capital among the different classes also determine the choice of alcoholic beverage. Bourdieu obtained similar results in France in the 1970s [Bourdieu 1984]; however, the types of beverages differ somewhat in Russia. The members of the highest classes in Russia prefer wine, champagne, cognac, whisky and exotic beverages such as rum and tequila. In addition, the consumption volume is not an indicator of luxury, as was the case with meat consumption in the Middle Ages [Braudel 1979]. On the contrary, the greatest volumes of beer and vodka, which are consumed by all classes, are consumed by the poorer and less educated. For these drinks and for wine, the most significant differences among classes occur for the choice of brands and manufacturers. The members of the lowest classes are focused on affordability and availability and tend to choose cheap Russian brands, whereas the members of the highest classes are focused on quality and taste and tend to choose more expensive foreign brands.
To conclude, in modern Russia there is a strong correlation between the types of alcohol consumed and social classes; Bourdieu reported similar findings in 1984. Features of the postmodern lifestyle, with its variegated consumption patterns (eclectic tastes in our case) were found in fewer than 5% of the alcohol consumers, most of whom belong to the educated and well-off groups of young people employed in executive positions. However, there is currently no reason to believe that these alcohol consumption patterns will spread from the rich to the poor, even though the trickle-down theory suggests that goods that were once considered luxury become routine [Braudel 1979]. Here, the eclectic consumption pattern is more likely an indicator of a high social status.
References
Andrienko Y., Nemtsov A. (2005) Estimation of Individual Demand for Alcohol. Economics Education and Research Consortium Working Paper Series, no 05/10.
Baudrillard J. (1968) Le système des objets [The System of Objects], Paris: Gallimard (in French).
Beck U. (1992) Risk Society: Towards a New Modernity, London: Sage.
Becker G., Murphy K. (1988) A Theory of Rational Addiction. Journal of Political Economy, vol. 96, no 4, pp. 675-700.
Bourdieu P. (1984 [1979]) Distinction. A Social Critique of the Judgment of Taste ftransl. by Richard Nice), London: Routledge & Kegan Paul.
Braudel F. (1979) Civilisation matérielle, économie et capitalisme, XVe - XVIIIe siècles [Civilization and Capitalism, 15th - 18th Century], 3 vols. Tome 1. Les Structures Du Quotidien: Le Possible et l'Impossible [Vol. 1: The Structures of Everyday Life: The Limits of the Possible], Paris: Armand Colin, Ch. 3 (in French).
Brennan A., Latimer N., Meng Y., Rafia R., Purshouse R. (2009) Modelling to Assess the Effectiveness and Cost-Effectiveness of Public Health Related Strategies and Interventions to Reduce Alcohol Attributable Harm in England Using the Sheffield Alcohol Policy Model Version 2.0. Report to the NICE Public Health Programme Development Group, 9 November 2009.
Clements K. W., Yang W., Zheng S. W. (1997) Is Utility Additive? The Case of Alcohol. Applied Economics, no 29, pp. 1163-1167.
Douglas M. (1987) A Distinctive Anthropological Perspective. Constructive Drinking: Perspectives on Drink from Anthropology (ed. M. Douglas), Cambridge: Cambridge University Press, pp. 3-15.
Douglas M., Isherwood B. (1979) The World of Goods: Towards an Anthropology of Consumption, New York: W. W. Norton.
Elias N. (1969) The Civilizing Process, Vol. I. The History of Manners, Oxford: Blackwell.
Elias N. (1982) The Civilizing Process. Vol. II. State Formation and Civilization, Oxford: Blackwell.
Farrell P., Fuchs V. (1982) Schooling and Health: The Cigarette Connection. Journal of Health Economics, no 1, pp. 217-230.
Grossman M., Sindelar J. L., Mullahy J., Anderson R. (1993) Policy Watch: Alcohol and Cigarette Taxes. Journal of Economic Perspectives, vol. 7, no 4, pp. 211-222.
Harvey D. (1989) The Condition ofPostmodernity: An Enquiry into the Origins of Cultural Change, Oxford, UK; Cambridge, MA: Basil Blackwell.
Herpen N., Verger D. (2008) Consommation et modes de vie en France, Paris: La Découverte, coll. "Grands Repères" (in French).
Hirschi T. (1998 [1969]) Social Bond Theory. Criminological Theory: Past To Present, Los Angeles: Roxbury.
Hughes K., MacKintosh A. M., Hastings G., Wheeler C. (1997) Young People, Alcohol, and Designer Drinks: Quantitative and Qualitative Study. British Medical Journal, no 7078, pp. 414-418.
Ionin L. (1998) Sociologiya kul'tury [Sociology of Culture], Moscow: Logos (in Russian).
Klimova S. (2007) Alkogolizm: obydennye teorii [Alcoholism: Ordinary Theories]. Social'naya real'nost' (electronic journal), no 2, pp. 30-40. Available at: http://corp.fom.ru/uploads/socreal/post-225.pdf (accessed on 11 January 2014) (in Russian).
Kozyreva P., Dorofeeva Z. (2008) Obshcheye i osobennoye v adaptacii gorozhan i sel'an k radikal'nym transformatsiyam [The Common and Particuliar Processes in Rural People's Adaption to Radical Changes]. In: Modernizaciya social'noy struktury rossijskogo obshchestva [Modernization of the Social Structure of the Russian Society] (ed. Z. T. Golenkova), Moscow: Institut sociologii RAN, pp. 69-96 (in Russian).
Levi-Strauss C. (1981 [1947]) Les stuctures élémentaires de laparenté [The Elementary Structures of Kinship], Paris: Mouton (in French).
Mennell S. (1985) All Manners of Food. Eating and Taste in England and France from the Middle Ages to the Present, Oxford: Basil Blackwell.
Mennell S. (1987) On The Civilizing of Appetite. Theory, Culture & Society, vol. 4, no 2-3, pp. 373-403.
Neufeld M., Rehm J. (2013) Alcohol Consumption and Mortality in Russia since 2000: Are There any Changes Following the Alcohol Policy Changes Starting in 2006? Alcohol and Alcoholism, vol. 48, no 2, pp. 222230.
Parker D., Brody J. (1982) Risk Factors for Alcoholism and Alcohol Problems among Employed Women and Men. In: Occupational Alcoholism: A Review of Research Issues. National Institute on Alcohol Abuse and Alcoholism, Research Monograph No. 8, Washington, DC: U.S. Government Printing Office, pp. 99127.
Peirce R. S., Frone M. R., Russel M., Cooper M. L. (1994) Relationship of Financial Strain and Psychosocial Resources to Alcohol Use and Abuse: The Mediating Role of Negative Affect and Drinking Motives. Journal of Health and Social Behavior, vol. 35, no 4, pp. 291-308.
Popova S., Rehm J., Patra J., Zatonski W. (2007) Comparing Alcohol Consumption in Central and Eastern Europe to Other European Countries. Alcohol & Alcoholism, vol. 42, no 5, pp. 465-473.
Roshchina Y. (2012) Dinamika i struktura potrebleniya alkogolya v sovremennoj Rossii [Dynamics and Structure of Alcohol Consumption in Modern Russia]. In: Vestnik Rossiyskogo monitoringa ekonomicheskogo polozheniya i zdorov'ya naseleniya (RLMS-HSE) [The Bulletin of Russian Longitudinal Monitoring Survey of Economic State and Health of Russian Population] (ed. P. M. Kozyreva), Moscow: HSE Publishing House, pp. 238-257 (in Russian).
Roshchina Y. (2013) To Drink or Not to Drink: The Microeconomic Analysis of Alcohol Consumption in Russia in 2006-2010. WP BRP 20/S0C/2013, Moscow: National Research University Higher School of Economics.
Shatikhin A. I. (2012) Pivnoy alkogolizm: nadumannaya problema ili novaya real'nost'? [Beer Alcoholism: An Unreal Problem or The New Reality?]. RusskiyMedicinskiy Zhurnal, vol. 15. Available at: http://www. rmj.ru/articles_8289.htm (accessed on 11 January 2014) (in Russian).
Tapilina V. (2006) Skol'ko p'et Rossiya? Ob'em, dinamika i differentsiatsiya potrebleniya alkogolya [How Much Do Russian People Drink? Volumes, Dynamics and Differentiation of Alcohol Consumption]. Sociologicheskie issledovaniya, no 2, pp. 85-94 (in Russian).
Thornton M. (1987) Sekt versus Schnapps in an Austrian Village. Constructive Drinking: Perspectives on Drink from Anthropology (ed. M. Douglas), Cambridge: Cambridge University Press, pp. 102-112.
WHO. 2012. European Action Plan to Reduce the Harmful Use of Alcohol 2012-2020, Geneva: World Health Organization.
