Impact of a New Alcohol Policy on Homemade Alcohol Consumption and Sales in Russia

Transcription

1 Alcohol and Alcoholism, 2015, 50(3) doi: /alcalc/agv008 Advance Access Publication Date: 25 February 2015 Article Article Impact of a New Alcohol Policy on Homemade Alcohol Consumption and Sales in Russia Vadim Radaev* Department of Sociology, National Research University Higher School of Economics, Myasnitskaya, 20, Moscow , Russia *Corresponding author: Tel.: ; Fax: ; radaev@hse.ru Received 23 October 2014; Revised 14 January 2015; Accepted 26 January 2015 Abstract Aims: To describe the effects of Russian policy since 2006 affecting price and availability on the consumption of recorded and unrecorded alcohol, with specific reference to homemade alcohol, and to investigate other factors affecting homemade alcohol consumption and purchasing. Methods: Consumption and preferred beverage data were collected from RLMS-HSE nationwide panel surveys from 1994 to 2013, with a detailed analysis of 2012 data (18,221 respondents aged 16+ years). Official statistics on manufactured alcohol sales, regional price increase and real disposable income were used. Results: Homemade distilled spirits (samogon) consumption decreased together with that of recorded and unrecorded manufactured spirits since The consumption of spirits was partially replaced by the consumption of beer and wine. These trends in alcohol consumption were interrupted in The interruption was more likely affected by the economic crisis and recession than by the new alcohol policy. Social networks and availability of unrecorded alcohol were more important predictors of homemade alcohol consumption and purchasing than was a recorded alcohol price increase. Conclusions: Homemade alcohol consumption does not replace the declining market for recorded spirits in Russia. The effects of economic and social factors on homemade alcohol consumption are greater than are the short-term effects of the new alcohol policy. The very recent (2015) reduction of the minimum unit price of vodka may be premature. INTRODUCTION Russia has been identified as a society with a high level of per capita alcohol consumption, prevalence of distilled spirits, and spread of heavy drinking (Haworth and Simpson, 2004; Popova et al., 2007). The country holds the fourth place in the world for per capita alcohol consumption, two and one-half times higher than the world average level (WHO, 2014). Combating excessive alcohol consumption, the Russian Government has been implementing a comprehensive reform, increasing alcohol taxation and imposing numerous restrictions on the manufacturing, selling and purchasing of alcoholic beverages since 2006 (Levintova, 2007; Nemtsov, 2011). In this paper, we address the issue of unrecorded alcohol consumption comprising 24% of total alcohol consumption in Russia (WHO, 2014), given that it is not often examined in the existing literature. This is especially true for post-communist countries (Lachenmeier et al., 2011). The effectiveness of Russian policies to reduce the consumption of unrecorded alcohol has been discussed in a few studies (Khaltourina and Korotayev, 2008; Gil et al., 2009; Nemtsov, 2009, 2011; Neufeld and Rehm, 2013). Additionally, existing research on the influence of illicit and non-beverage alcohol has largely focused upon alcoholrelated harm and its effects on health and mortality (Leon et al., 2009; Solodun et al., 2011; Dissing et al., 2013). However, economic conditions and drinking patterns of unrecorded alcohol consumption have been less studied. We focus on homemade alcohol as a category of unrecorded alcohol (Nordlund and Osterberg, 2000; Rehm et al., 2010). It includes non-commercial alcohol made for in-house consumption and commercial alcohol produced for sale. In previous chemical studies, The Author Medical Council on Alcohol and Oxford University Press. All rights reserved 365

2 366 Alcohol and Alcoholism, 2015, Vol. 50, No. 3 when correct comparison of unrecorded alcohol with compatible forms of recorded alcohol was taken, the quality of homemade spirits was qualified even higher than quality of legally manufactured spirits (Rehm et al., 2014). However, commercial homemade alcohol produced for the illegal market and aimed at ensuring intoxication might be exposed to higher risks and more harmful to consumer health (Nemtsov, 2009). Relationships between alcohol taxation, alcohol prices and its consumption have been thoroughly studied (Chaloupka et al., 2002; Mäkelä et al., 2007; Wagenaar et al., 2009). Normally, alcohol is treated as a standard commodity with relatively high price elasticity. A decrease of alcohol consumption resulting from the increase in alcohol prices was claimed to be a stable result in the economic literature in Russia and worldwide (Andrienko and Nemtsov, 2005). A metaanalysis of 1003 estimates from 112 studies demonstrated that excise and sales taxes were passed through to prices, typically at passthrough rates of one to two. There was also strong statistical evidence of the effects of alcohol prices on the drinking of all beverage types (Wagenaar et al., 2009); although with regard to Russia, some social factors were recognized as more important than the price of alcohol (Roshchina, 2013). At the same time, the effects on the use of unrecorded alcohol of price increases on recorded alcohol were rarely considered. Previous historical evidence was diverse in Russia (Nemtsov, 2009). However, where homemade or other unrecorded alcohol was readily available, its consumption was likely to increase because of a price increase (Haworth and Simpson, 2004). In particular, substitution of vodka for moonshine is expected when price of vodka grows (Andrienko and Nemtsov, 2005). Dealing with unrecorded and recorded commercial beverages may require different policy tools. Addressing one type of beverage may impose unintended effects on another. For example, strict government controls over legal alcohol generally lead to a decrease of total alcohol consumption but may provide a shift to the unrecorded market (Thamarangsi, 2013). The structure of this paper is the following. We outline a new alcohol policy in Russia. Statistical and survey data are used to reveal trends in recorded and homemade alcohol consumption. Two groups of consumers dealing with homemade distilled spirits (samogon) and homemade wines are delineated. Factors affecting consumption and purchasing of homemade alcoholic beverages are investigated. We conclude that the short-term outcomes of the new alcohol policy in Russia are not clear and are outweighed by social factors. Combating alcohol in Russia: historical background Continuous attempts have been made by the Russian authorities to restrain alcohol consumption and combat the illicit alcohol market since the eve of the Soviet era, when manufacture, selling, and purchasing of samogon were banned and became subject to criminal prosecution in However, enforcement was not very effective, and samogon prevailed over legal alcoholic beverages in the 1920s. The ratio of samogon to vodka was estimated as four to one, with 10% of all households producing and selling homemade distilled spirits (Zaigraev, 2004). After a sharp decline of both recorded and unrecorded alcohol during World War II, there were several waves in the development of samogon consumption in Russia. These recurrent cycles each lasted for years (Skog, 1986). During the rising wave in the post-war period, samogon reached 53% of total alcohol consumption in Further growth of legal alcohol consumption was accompanied by a continuous decline of samogon consumption, decreasing to 27% in (Treml, 1982, 1997; Nemtsov 2009). Anti-alcohol campaigns of 1958 and 1972 contributed to this decrease. Gorbachev s anti-alcohol reform of brought significant alcohol price increases, restrictions on alcohol sales, and heavy penalties for alcohol-related offences. As a result, use of legal alcohol was abruptly cut by >50% (White, 1996; Bhattacharya et al., 2012). At the same time, samogon consumption increased by 80%, accounting for 64% of the total alcohol consumption in At least onehalf of the diminishing recorded alcohol was replaced by unrecorded alcohol (Vroublevsky and Harwin, 1998; Nemtsov, 2009). After the end of Gorbachev s campaign, in consumption of legal alcohol partially recovered and samogon contributed 54% to the total alcohol consumption. Economic reforms in brought deregulation of trade accompanied by dramatic price increases and massive substitution of vodka for unrecorded alcohol (Andrienko and Nemtsov, 2005). Overall consumption of samogon in the 1990s increased (Tapilina, 2006). However, the volume of unrecorded alcohol in that period grew mainly due to the tide of cheap and largely illegal spirits (including counterfeited beverages and pure ethanol) imported and smuggled to Russia by the new private entrepreneurs (Nemtsov, 2009). The same inflow of imported alcohol was observed at that time in the other Eastern European countries (Moskalewicz and Simpura, 2000). Finally, the economic growth of the 2000s stimulated increased recorded alcohol and decreased unrecorded alcohol consumption. A new alcohol policy was enacted in Russia in 2006, when tax stamps were introduced together with electronic movement and surveillance systems to track the alcohol market and alcohol movement (Levintova, 2007; Gil et al., 2009). An active phase of the reform was started after 2009, when the Federal Service for Alcohol Market Regulation (Rosalkogolregulirovanie) was established. A wide range of new regulations was imposed on the manufacture and sale of alcoholic beverages. An accelerated increase of alcohol excise taxes occurred, reaching 30% in and 80% in Minimum unit prices on vodka were raised by two and one-half times (from 89 to 220 RBL per 0.5 l) in The resulting retail price of vodka increased by 77% in , whereas the price of dry wine grew by 28% but the real disposable income of the population increased by only 15%. Restrictions on alcohol availability included bans on off-premise sales of alcoholic beverages in the evening and at night. Points of sale had to be located at a safe distance from educational, medical, sports, and cultural facilities. A ban on alcohol consumption in public settings was added. Step-by-step limitations and finally a total ban on above-the-line advertising of alcoholic beverages were enacted. With regard to homemade alcohol, some experts suggested a complete ban on home distilling (Khaltourina and Korotayev, 2008), although it was unclear how such a ban could be enforced. METHOD Data were collected from the Russia Longitudinal Monitoring survey, RLMS-HSE, established by Demoscope and the Carolina Population Center, University of North Carolina at Chapel Hill in 1992 and jointly conducted with the Higher School of Economics since RLMS-HSE includes a series of annual nationwide panel surveys. Publicly available data represent the Russian population aged 14+ years from 1994 to Self-reports of drinking incidence, volume of drinking, and beverage preferences were the main measures. We used ratios of legal and

3 Alcohol and Alcoholism, 2015, Vol. 50, No homemade alcohol consumers to investigate trends in consumption of different beverages. Self-reports are backed by the official Rosstat statistics on sales of recorded and unrecorded alcohol in Unrecorded manufactured spirits consumption is calculated as the difference between official sales and production of vodka and hard liquors. To examine homemade alcohol consumption and purchasing in more detail, we used the RLMS-HSE 2012 round data collected from adults (16+). The main sampling parameters are summarized in Table 1. Survey data on homemade alcohol included three beverage types: samogon, homemade wines and homebrew (braga). Samogon has been traced in the RLMS-HSE surveys since Braga has been tracked separately from beer since Questions concerning homemade wine were first added to the survey in Beverage preferences were investigated with regard to consumed and purchased homemade alcohol. We ran three binary logistic regressions using incidence of samogon consumption, homemade wine consumption, and purchasing of homemade alcohol as dependent variables. The increase of average regional prices on vodka and wine during the active phase of alcohol reform in is used to measure a new policy effect. The average per capita income (natural logarithm) in 2012 and the increase of real disposable income of the population in the surveyed regions in are used as control variables. Alcohol availability in the localities at night is used as a proxy for availability of unrecorded alcohol (both manufactured and homemade), given that sales of recorded alcohol at night is prohibited by law. Table 1. Main sampling parameters (n = 18221, %) Male 42.0 Female 58.0 Age Education Primary school 18.8 Secondary school 56.3 University degree 24.1 Marital status Married (live together) 62.8 Non-married 37.2 Employed 57.6 Non-employed (or work <8 h a week) 42.4 Type of residence (number of inhabitants) >1 mln thd thd thd 10.2 <100 thd 24.0 Village 25.0 Regions of Russia North-West 9.2 Central 18.7 South 16.2 Povolzhie 19.7 Ural 9.0 Siberia and Far-East 17.8 Moscow 9.5 Among other predictors, we differentiate between Southern and Northern regions of Russia and distinguish urban and rural areas. Assuming that social networks may encourage or discourage the use of homemade alcohol, we first asked our respondents whether they had any relatives, neighbours or acquaintances producing homemade alcoholic beverages and, second, if they had acquaintances consuming non-beverage alcohol (colognes, lotions, pharmaceutical alcoholic tinctures and technical fluids). The economic status of respondents was measured by their personal income (natural logarithm) and the number of durable goods accumulated at their households. Gender, age, educational level and employment status were added. The main variables and measures are summarized in Table 2. To check the robustness of the obtained results, we ran the same regression for the subsample of respondents drinking alcohol during the last 30 days and the subsample of all alcohol users including those who did not drink during the last 30 days. We obtained similar results, proving that the inclusion of all alcohol users who consumed alcoholic beverages at least sometimes is not critical for the model. The same calculation was conducted for buyers and non-buyers of homemade alcohol. RESULTS Decline of recorded and unrecorded distilled spirits According to Rosstat data, the adult per capita consumption of recorded vodka and other spirits measured in litres of pure ethanol Table 2. Main variables and measures Variables Use of alcoholic beverages Prices for recorded alcohol Availability of unrecorded alcohol Regional differentiation Type of residence Social networks Gender Age Education Employment Income level Accumulated wealth Measurers Drinking alcoholic beverages during the last 30 days (including preferred beverage types) Purchasing alcoholic beverages during the last 30 days (including preferred beverage types) Regional average prices of vodka and wine a Regional price increase for vodka and wine in a Availability of alcohol for purchasing at night time Southern vs. other regions b Urban vs. rural residence Cultivate private land plots Have acquaintances, neighbours, or relatives who during the last 30 days: make home-produced alcoholic beverages consume alcohol surrogates c Females vs. males Number of years Have or do not have university degree Employed vs. non-employed Personal income during the last 30 days (all sources) d The number of durable goods possessed by the household e a Rosstat statistics is used a source of data. b Krasnodar, Stavropol and two North-Caucasian regions are attributed to the first cluster of Southern regions and the other regions to the second cluster. c We did not ask the respondents directly if they consume alcohol surrogates themselves due to the high sensitivity of this question. d Natural logarithm of income is used. e Durables were selected by the respondents from the list of 37 items.

4 368 Alcohol and Alcoholism, 2015, Vol. 50, No. 3 Fig. 1. Recorded consumption of vodka and spirits, beer, dry wine, and unrecorded spirits in Russia in , per capita, in litres of pure ethanol. Source: Rosstat. Note: Unrecorded spirits have been calculated by Irina Kratko (Higher School of Economics) as difference between official sales and production of vodka and spirits corrected by export/import balance. Data on 2013 is not available. was declining with some fluctuations throughout the entire observation period of Unrecorded manufactured spirits consumption also declined in Both recorded and unrecorded spirits were partially replaced by the consumption of beer and wine until 2007, with an interruption of trends afterwards (Fig. 1). Turning to the RLMS-HSE survey data, we observe that the percentage of drinkers of vodka and other spirits during the last 30 days was declining during , stabilized in , and decreased further in The percentage of beer drinkers increased in and stabilized in , implying that the same number of drinkers consumed an increasing volume of beer. Then, both parameters decreased after The percentage of wine drinkers after a downward trend in changed to an upward trend in , with a stabilization afterwards. The percentage of samogon drinkers grew rapidly, starting from a relatively low level in Since 2001, we observed a downward trend from 19 to 5% of drinkers that, remarkably, halted in Males consumed samogon much more often than did females over all periods of observation. Younger groups were always less involved in samogon consumption. The percentage of samogon drinkers among rural dwellers with scarce pecuniary resources was normally 2-fold greater than average. However, the shape of the curve remained the same for all groups under study. This implies that self-reported unrecorded samogon consumption declined since 2001 in parallel with recorded and unrecorded manufactured spirits measured by objective statistics (Fig. 2). In previous studies, a clear downward trend in recorded and unrecorded spirit consumption in the 2000s was interpreted as a partial success of the new governmental alcohol policy (Neufeld and Rehm, 2013). We suggest a slightly different interpretation, relying upon both statistical and survey data. Almost all major trends in the consumption of recorded and unrecorded alcohol were observed in , when annual growth rates of GDP and real disposable income of the population reached 7 and 10%, respectively. These trends were interrupted during the financial crisis in and afterwards when GDP annual growth rates declined from 4 to 1% in Average real disposable income growth rates accounted for 3% in The new alcohol policy measures could have contributed to the observed changes. However, we emphasize that the interruption of trends happened before the accelerated growth of excise taxes and minimal unit prices in Thus, the interruption was more likely affected by the crisis of and the following recession, when consumers reduced alcohol consumption or moved to cheaper alcoholic beverages including unrecorded alcohol. Consumption and purchase of homemade alcohol Starting with the basic facts from the 2012 round of the RLMS-HSE survey, 69.4% of respondents consumed alcoholic beverages at least sometimes. Of these, 71.3% of drinkers (49.3% of all respondents) consumed alcoholic beverages during the last 30 days. Data showed that 9.9% of drinkers (4.9% of all respondents) consumed homemade alcoholic beverages during the last 30 days including 4.7% samogon consumers, 5.2% homemade wine consumers and 0.6% homebrew consumers (for descriptive statistics, see Table 3). Homemade alcoholic beverages constituted 10.7% of the overall volume of consumed alcohol measured in grams of pure ethanol. Samogon was still prevalent, accounting for 81.2% of overall homemade alcohol volume, whereas homemade wines and homebrew constituted 17.7 and 1.1%, respectively. The share of samogon in the total volume of consumed ethanol accounts for 8.7%. It is approximately one-fifth of the self-reported volume of consumed vodka, equalling 41.3% of the total consumed alcohol. This is consistent with the official Rosstat statistical data. We do not consider homebrew in this paper, given that its consumers constitute only % of drinkers. As for consumers of samogon and

5 Alcohol and Alcoholism, 2015, Vol. 50, No Fig. 2. Percentage of recorded alcohol and samogon drinkers during the last 30 days in , RLMS-HSE data. Table 3. Consumption of homemade alcoholic beverages by groups of respondents, n = (8935), % Samogon Homemade wine Total sample 3.3 (4.7) 3.7 (5.2) Male 5.2 (6.6) 3.0 (3.8) Female 1.7 (2.6) 4.4 (6.8) Age (0.9) 3.1 (5.2) (2.7) 2.5 (3.4) (3.6) 3.4 (4.5) (5.3) 4.5 (5.9) (6.1) 4.2 (6.0) (8.2) 4.7 (8.2) Education Primary 6.8 (10.2) 2.8 (4.2) Secondary 3.4 (4.7) 3.6 (5.0) Higher 1.2 (1.7) 4.6 (6.3) Employed 2,3 (3.1) 3.7 (4.9) Non-employed (or work <8 h a week) 5.2 (8.3) 3.8 (6.0) Regional differences Southern regions 1.7 (2.8) 7.2 (11.7) Other regions 3.5 (4.9) 3.3 (4.5) Type of residence Urban areas 2.3 (3.2) 3.8 (5.2) Rural areas 6.9 (10.1) 3.6 (5.3) Social networks Have acquaintances producing 9.7 (12.6) 10.5 (13.6) homemade alcohol Have acquaintances consuming alcohol surrogates 8.0 (10.6) 2.2 (2.9) Note: In parentheses drinkers only (consuming alcoholic beverages during the last 30 days). homemade wines, these groups almost do not overlap. Only 3% of homemade alcohol consumers drink both samogon and homemade wine, implying that these two groups are almost entirely different. Since 2001, the production of homemade alcohol has been allowed for personal use, but sale is prohibited as before (Volkov et al., 2012, p. 49). Nevertheless, some homemade alcohol is sold on the illegal market. It is not easy to address this issue from the supply side because producers are reluctant to report sales of homemade alcohol. Turning to the demand side, we asked our respondents if they purchased any alcoholic beverages, homemade alcohol in particular, during the last 30 days. Alcoholic beverages were purchased by 50% of alcohol drinkers (25% of all respondents). In turn, 6.5% of alcohol buyers (1.6% of all respondents) reported that they purchased homemade alcoholic beverages. Most of these dealt with samogon (88%); purchasing of other beverages was rare. Thirty-seven per cent of samogon drinkers purchased it during the last 30 days. The others produced homemade alcohol themselves or obtained it from their friends as a gift. Factors affecting homemade alcohol consumption The incidence of samogon and homemade wine drinking and the purchasing of homemade alcohol during the last 30 days were used as three dependent variables in binary logistic regression models. Regression coefficients are presented in Table 4. We used the increase of average prices on vodka and wine in in the surveyed regions to indicate an impact of the new alcohol policy. Surprisingly, we obtained a significant negative effect of the price increase of vodka on percentage of samogon consumers and buyers (P < 0.001). The increase of average regional prices of manufactured wine was lower, and its effect on consumption of homemade wine was insignificant. Our explanation is the following. Regions with lower population per capita income were more involved in samogon consumption (P < 0.01), whereas the increase of real disposable income in the surveyed regions in was negatively related to samogon consumption (P < 0.01). The vodka price increase in was higher in regions with higher levels of population per capita income. This implies that regions consuming more samogon were less affected by the price increase.

6 370 Alcohol and Alcoholism, 2015, Vol. 50, No. 3 Table 4. Binary logistic regression, dependent variables are consumption of samogon, consumption of homemade wine, and purchasing homemade alcohol during the last 30 days (RLMS-HSE, 2012) Consume samogon Consume homemade wine Purchase homemade alcohol Increase of regional prices on manufactured vodka/wine in *** (0.009) (0.009) 0.926*** (0.010) Availability of alcohol at night time 1.514* (0.167) 1.366* (0.146) 1.771** (0.204) Southern regions 0.306*** (0.269) 3.915*** (0.185) 0.204*** (0.316) Villages 1.537** (0.154) (0.168) (0.185) Have acquaintances producing homemade alcohol 5.626*** (0.144) 3.798*** (0.141) *** (0.179) Have acquaintances consuming alcohol surrogates 1.879*** (0.152) 0.487** (0.239) 3.783*** (0.170) Gender (females) 0.309*** (0.158) 1.577** (0.141) 0.529*** (0.171) Age 1.017** (0.005) 1.018*** (0.005) (0.006) Higher education (0.229) (0.156) (0.267) Employment 0.701* (0.176) (0.182) (0.207) Personal income (Ln) 0.760** (0.091) 0.842* (0.087) 0.777* (0.101) Number of durables 0.913*** (0.020) 1.051** (0.019) 0.899*** (0.024) Regional average per capita income, ** (0.302) (0.230) (357) Increase of regional real disposable income in (Ln) 0.964** (0.013) 0.968* (0.012) (0.015) Constant *** (3.105) (2.348) (3.647) R N Notes: standard errors in parentheses. *P < **P < ***P < (two-tailed tests). Availability of alcohol for purchasing at night was reported by 38% of respondents. It was positively related to the ratio of both samogon and homemade wine consumers (P < 0.05). The probability of consuming homemade alcohol increased by times in localities where unrecorded alcohol was always available. Similar results were obtained for purchasing behaviour. If alcohol was available at night, people bought homemade alcoholic beverages in the market more often (P < 0.01). Living in the Southern regions of Russia had a pronounced effect on the consumption of homemade alcohol that could be attributed to climate and culture. Association was positive for the consumption of homemade wine and negative for the consumption and purchasing of samogon (P < 0.001). Rural dwellers consumed samogon more frequently than urban dwellers (P < 0.01). This association was insignificant for homemade wine consumption and the purchase of homemade alcohol. Social networks were the most important predictor of homemade alcohol consumption and purchase. If people had relatives, neighbours or acquaintances producing homemade alcoholic beverages (14% of drinkers), the probability of homemade alcohol consumption increased 6-fold for samogon and 4-fold for homemade wine. This effect size was the largest among all covariates. It was especially large in the case of homemade alcohol purchase (P < 0.001), implying that homemade alcohol was bought primarily through network ties. According to self-reports, 63% of homemade alcohol buyers obtained it from their neighbours. In our sample, 13% of drinkers had acquaintances consuming non-beverage alcohol. These hazardous network ties increased the probability of samogon drinking and purchasing (P < 0.001) and decreased the probability of homemade wine drinking (P < 0.01). The gender impact on consumption was significant and diverse. Men were more frequently involved in the consumption and purchasing of samogon (P < 0.001), whereas homemade wine was more popular among women (P < 0.01). The age of respondents was positively related to the consumption of homemade alcoholic beverages. The effect size was small but significant, implying that older people tended to drink homemade alcohol more often. These results were valid both for samogon (P < 0.01) and homemade wine drinkers (P < 0.001). No significant association was revealed with purchasing behaviour. Educational level did not have a significant effect on the use of homemade alcohol. Unemployed people were overrepresented among samogon drinkers and buyers (P < 0.05). This difference was insignificant for homemade wine. Personal income was negatively related to the propensity to consume both samogon (P < 0.01) and homemade wine and to purchase homemade alcohol (P < 0.05). Regarding accumulated wealth, people with a larger number of durable goods at their households tended to avoid drinking and purchasing samogon (P < 0.001), but the probability of using homemade wine was slightly greater (P < 0.01). DISCUSSION Rosstat data on changes in beverage preferences are consistent with the trends pointed out in the WHO Russia s country profile (WHO, 2014). The objective statistics are also largely consistent with self-reports collected from RLMS-HSE surveys in Both statistical and survey data demonstrated that homemade alcohol consumption stopped replacing the declining recorded spirits in Russia since Samogon consumption decreased together with recorded and unrecorded manufactured spirits. All spirits were partially replaced by the consumption of beer and wine at least until These trends started before the new alcohol reform and were driven largely by changes in GDP and real disposable income of the population in the 2000s, despite the fact that the first policy interventions occurred in Interruption of these trends during could have been influenced by restrictive policy measures, especially by the accelerated price increase on recorded alcohol in However, they could even more likely have been affected by the financial crisis of and the following recession, when consumers reduced alcohol consumption or moved to cheaper manufactured or homemade alcoholic beverages.

7 Alcohol and Alcoholism, 2015, Vol. 50, No We must admit that it is not easy to delineate a precise role of the policy intervention. Similar difficulties were observed, for example, in Estonia as a former socialist country. Estonia had a level of drinking similar to that of Russia. Similar restraining measures were undertaken from the mid-2000s. Still, it is currently unclear how much of the subsequent decline in alcohol consumption in Estonia was caused by policy changes and how much was an effect of the economic downturn measured by per capita GDP and the affordability of alcohol (real disposable income/prices) (Lai and Habicht, 2011). Cultural shifts could also affect changes in consumption patterns by stimulating substitution effects. For example, younger people increasingly turned from vodka to beer, and more educated people turned from spirits to wine in Russia in the 2000s (Denisova, 2010). These changes could develop according to inner rules driven by longrange indirect network ties (Skog, 1986). In any case, short-term effects of alcohol reform are intertwined with the influence of other exogenous and endogenous factors, with some delayed effects that may become fully effective in the future. Regarding the consumption of homemade alcohol, we distinguish groups of samogon and homemade wine consumers that are similar in size but demonstrate divergent drinking patterns. Samogon prevails in volume of consumed alcohol. Most people purchasing homemade alcohol in the illegal market also consume distilled spirits. Contrary to expectations, the price increase on recorded alcohol in the surveyed regions demonstrates no positive influence on the incidence of homemade alcohol consumption. Their association is mediated by per capita income differentiation and increase. People in the regions with lower per capita income were more involved in samogon consumption. At the same time, these regions were less affected by the price increase for vodka. The availability of alcoholic beverages for purchasing at night is positively related to the use of homemade alcohol. The state should impose more-effective controls to curb the distribution and sales of commercial unrecorded alcohol. The character of social networks plays a remarkable role in predicting homemade alcohol consumption and purchase. It is consistent with previous findings that peers play a significant role in decisionmaking regarding drinking (Yakovlev, 2012). Both samogon and homemade wine consumers are attached to local networks of homemade alcohol producers. Samogon consumers more often have neighbours consuming non-beverage alcohol, whereas wine consumers maintain a safe distance from users of alcohol surrogates. Homemade alcohol consumers are older than the average population and have lower personal income. Samogon consumers and buyers could be found more easily in rural areas. Males and the unemployed are overrepresented among samogon users. Consumers of homemade wine are represented more by females and urban dwellers located in the Southern regions of Russia. Overall, social factors are more important than the price of alcohol (Roshchina, 2013). This study is limited by the analysis of homemade alcohol and does not consider illegally manufactured alcohol and non-beverage alcohol that is legally produced but not officially intended for human consumption. Surrogate alcohol poses the highest risk of fatal poisoning (Gil et al., 2009; Solodun et al., 2011). It is also an important limitation of this study that we address short-term effects of the new alcohol reform. We agree that it may be too early for conclusions regarding the success of governmental attempts to curb alcohol consumption (Neufeld and Rehm, 2013). The actual effects of increases in excise taxes and minimum unit prices of spirits remain largely unclear. Policy measures could produce delayed effects on a continuous process of alcohol consumption. However, despite the absence of strong empirical evidence, the Russian government perceived a decline of the legal alcohol sales in as an immediate negative effect of the new alcohol policy and started to correct this policy in Instead of further increase of excise taxes on spirits by 32% which was planned in , excise tax rates were frozen for these years. The minimal unit price on vodka was decreased by 16% from 220 to 185 RBL per 0.5 l in It was the first decrease of this price since its introduction in TV and radio advertising of wine banned in 2012 was allowed again for domestic wine since January It would be premature to claim that the new alcohol policy is reversed but some important restrictive measures have been turned down. A necessity to soften alcohol policy is publicly explained by the expansion of illegal alcohol markets due to the price increase on legal alcohol. We would insist that most recent policy developments run ahead of time. Changes in the structure of alcohol markets and consumption patterns are a highly complex process affected by a constellation of exogenous and endogenous factors. The contribution of these factors needs further exploration. A longer time span should be considered. New RLMS-HSE rounds data, once available, should prove helpful for investigating trends in homemade alcohol usage and revealing delayed effects of the policy measures. ACKNOWLEDGEMENTS We are thankful to Zoya Kotelnikova (HSE), Irina Kratko (HSE), Marjana Martinic (IARD), Yana Roshchina (HSE) and Anna Sheveland (IARD) for valuable comments on the draft of this paper. FUNDING The work was supported by an unrestricted grant for academic research from The International Center for Alcohol Policies (ICAP) (International Alliance for Responsible Drinking, IARD since 1 January 2015). RLMS-HSE surveys were funded by the Program for Basic Research of the National Research University Higher School of Economics. ETHICAL STATEMENT The RLMS-HSE study met the standards for the ethical treatment of participants. It was approved by the Institutional Review Boards of the University of North Carolina at Chapel Hill, No Monitoring the Social Safety Net in Russia, renewal approved 06/02/2014. CONFLICT OF INTEREST STATEMENT None declared. REFERENCES Andrienko Y, Nemtsov A. (2005) Estimation of individual demand for alcohol. Economics Education and Research Consortium Working Paper Series 05/10. Bhattacharya J, Gathmann C, Miller G. (2012) The Gorbachev anti-alcohol campaign and Russia s mortality crisis. IZA Discussion paper Chaloupka FJ, Grossman M, Saffer H. (2002) The effects of price on alcohol consumption and alcohol-related problems. Alcohol Res Health 26: Denisova I. (2010) Potrebleniye alkogolya v Rossii: Vliyaniye na zdorovie i smertnost [Consumption of alcohol in Russia: Effect on health and mortality]. CEFIR/NES Analytical Reports and Papers 31. DissingAS,GilA,KeenanK,et al. (2013) Alcohol consumption and selfreported (SF12) physical and mental health among working-aged men in a typical Russian city: a cross-sectional study. Addiction 108:

8 372 Alcohol and Alcoholism, 2015, Vol. 50, No. 3 Gil A, Polikina O, Koroleva N, et al. (2009) Availability and characteristics of nonbeverage alcohols sold in 17 Russian cities in Alcohol Clin Exp Res 33: Haworth A, Simpson R. (2004) Moonshine Markets: Issues in Unrecorded Alcohol Beverage Production and Consumption. New York: Brunner- Routledge. Khaltourina DA, Korotayev AV. (2008) Potential for alcohol policy to decrease the mortality crisis in Russia. Eval Health Prof 31: Lachenmeier DW, Taylor BJ, Rehm J. (2011) Alcohol under the radar: do we have policy options regarding unrecorded alcohol? Int J Drug Policy 22: Lai T, Habicht J. (2011) Decline in alcohol consumption in Estonia: combined effects of strengthened alcohol policy and economic downturn. Alcohol Alcohol 46: Leon DA, Shkolnikov VM, McKee V. (2009) Alcohol and Russian mortality: a continuing crisis. Addiction 104: Levintova M. (2007) Russian alcohol policy in the making. Alcohol Alcohol 42: Mäkelä P, Bloomfield K, Gustafsson NK, et al. (2007) Changes in volume of drinking after changes in alcohol taxes and travellers allowances: results from a panel study. Addiction 103: Moskalewicz J, Simpura J. (2000) The supply of alcoholic beverages in transitional conditions: the case of Central and Eastern Europe. Addiction 95: S Nemtsov AV. (2009) Alkogolnaya istoriya Rossii: Noveishii period [Alcoholic History of Russia: The new Times]. Moscow: Librocom. Nemtsov AV. (2011) A Contemporary History of Alcohol in Russia. Stockholm, Sweden: Södertörns Högskola. 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 Alcohol 48: Nordlund S, Osterberg E. (2000) Unrecorded alcohol consumption: its economics and its effects on alcohol control in the Nordic countries. Addiction 95: S Popova S, Rehm J, Patra J, et al. (2007) Comparing alcohol consumption in Central and Eastern Europe to other European countries. Alcohol Alcoholism 42: Rehm J, Kanteres F, Lachenmeier DW. (2010) Unrecorded consumption, quality of alcohol and health consequences. Drug Alcohol Rev 29: Rehm J, Kailasapillai S, Larsen E, et al. (2014) A systematic review of the epidemiology of unrecorded alcohol consumption and the chemical composition of unrecorded alcohol. Addiction 109: Roshchina Y. (2013) To Drink or not to Drink: The Microeconomic Analysis of Alcohol Consumption in Russia in WP BRP 20/SOC/ Moscow: National Research University Higher School of Economics. Skog OJ. (1986) The long waves of alcohol consumption: a social network perspective on cultural change. Soc Networks 8:1 32. Solodun Y, Monakhova Y, Kuballa T, et al. (2011) Unrecorded alcohol consumption in Russia: toxic denaturants and disinfectants pose additional risks. Interdiscip Toxicol 4: Thamarangsi T. (2013) Unrecorded alcohol: significant neglected challenges. Addiction 108: Tapilina V. (2006) Skolko piet Rossiya? Obyem, dinamika i differentsiatsiya potrebleniya alkogolya [How much does Russia drink: The volume, dynamics, and differentiation of alcohol consumption]. Sotsiologicheskiye Issledovaniya 2: Treml VG. (1982) Alcohol in the USSR. A Statistical Study. Durham, NC: Duke Press Policy Studies. Treml VG. (1997) Soviet and Russian statistics on alcohol consumption. In: Bobadilla JL, Costello CA, Mitchell F (eds). Premature death in the New Independent States. Committee on Population, Commission on Behavioral and Social Sciences and Education, National Research Council. Washington, DC: National Academy Press, Volkov AV, Kholdin VN, Gorbatchev IA, et al. (2012) Russia. In: Producers, Sellers, and Drinkers: Studies of Noncommercial Alcohol in Nine Countries. Washington, DC: ICAP, Vroublevsky A, Harwin J. (1998) Russia. In: Grant M. (ed). Alcohol and Emerging Markets: Patterns, Problems, and Responses. Philadelphia: Brunner and Mazel, Wagenaar AC, Salois MJ, Komro KA. (2009) Effects of beverage alcohol price and tax levels on drinking: a meta-analysis of 1003 estimates from 112 studies. Addiction 104: White S. (1996) Russia Goes Dry: Alcohol, State and Society. New York: Cambridge University Press. World Health Organization. (2014) Global Status Report On Alcohol and Health. Geneva: World Health Organization. Yakovlev E. (2012) Peers and Alcohol: Evidence from Russia. CEFIR/NES Working Paper series 182. Zaigraev G. (2004) The Russian model of noncommercial alcohol consumption. In: Haworth A, Simpson R (eds). Moonshine Markets: Issues in Unrecorded Alcohol Beverage Production and Consumption. New York: Brunner-Routledge,