Drinking Culture and Stock Returns in Europe

(1)

Drinking Culture and Stock Returns in Europe

Thesis

Master of Science in Business Administration

University of Groningen Faculty of Economics and Business

Abstract: This paper describes the influence of a drinking culture on the risk-adjusted stock returns of European companies. The performance of stocks of alcohol producing companies are examined for the European region, by employing the CAPM, the three-factor model and the four-factor model in a time-series analysis. Different drinking cultures are explained, which consist of dry, central and wet cultures. The main finding is that stocks of alcohol producing companies in central Europe have annually higher abnormal risk-adjusted returns of 0.36% compared to ordinary stocks within the same region. This finding is not found in countries who have a morally neutral attitude towards alcohol.

Keywords: Drinking culture, alcohol, social norms, excess returns

Author: Chris Gossink Student number: 1703595

Supervisor: prof. dr. Boudewijn P. de Bruin 2nd_{Supervisor: dr. Raymond Zaal}

(2)

1

List of tables

Table 1: Differences between wet and dry drinking cultures ... 7

Table 2: Equity home bias ... 10

Table 3. Regions of drinking cultures ... 12

Table 4. Country statistics ... 14

Table 5. Regression results for equally-weighted portfolios of dry, central and wet countries ... 17

(3)

2

1. Introduction

During the last decade, sin stocks, stocks of companies involved with vices such as the production of alcohol, tobacco, gambling, weaponry and adult entertainment, have received much attention from researchers, since a number of investors as well as fund managers shun these type of stocks because they do not accord with their social and ethical beliefs. Investors and fund managers who invest following the Sustainable and responsible investing (SRI) guidelines use, among other techniques, the screening of religious or moral vices, such as alcohol, tobacco and gambling, from portfolios. According to the US Sustainable and Responsible Investment Forum (2012), more than 11% of the investments under professional management in the United States was invested according to SRI strategies and they further estimate that SRI is likely to grow significantly the coming years.

Research has found these sin stocks to earn risk-adjusted abnormal returns in the US compared to other stocks because of the existence of social norms (Hong and Kacperczyk, 2009; Kumar and Page, 2011); research is scarce on what influences these norms or how to measure them. Some researchers try to explain the social norms or the more economical term “the economic attitudes of individuals, groups and societies” to religion (Salaber, 2013); others to political values, i.e. being a republican or democrat (Hong & Kostovetsky, 2012).

(4)

3 influences alcohol related deaths (Lelbach, 1976; Norström, 2001; Ramstedt, 2001, 2002; Rossow, 2001; Skog, 2001). Third, when taking all possible sins in consideration (see Renneboog et al., 2008 for a complete list) alcohol stocks make up a large proportion of the sample in terms of publicly listed companies. For example, when researching whether religion has an effect on sinful companies Salaber (2013) uses both alcohol and tobacco stocks in her sample, of which 59 stocks are alcohol producing companies and only 7 stocks are involved with the production of cigarettes. Because of these reasons this paper is focusing on the companies involved with the production and distribution of alcohol, these stocks and sin stocks are used interchangeably. In order to describe the culture with respect to alcohol in a country, some researchers have divided countries in either “wet” or “dry” countries (Allamani, Beck, Bergmark, Csemy, Eisenbach-Stangl, Elekes […] & Mendoza, 2006; Allamani, Voller, Kubicka & Bloomfield, 2000; Skog, 1985a Skog, 1985b; Ramstedt, 2001). Wet countries have a culture in which alcohol is part of everyday live. These citizens prefer a glass of wine with their meal and mildly consuming alcohol is considered normal, especially in a social setting. People from wet countries have a morally neutral view towards alcohol (SIRC, 1998). Dry countries on the other hand have a history of temperance movements and have more teetotalers than the wet countries and drinking is kept more secretive. They use alcohol as a way to intoxicate themselves (Leifman, 2002), instead of mildly consuming, the result is a higher percentage of alcohol poisoning cases. Dry countries face the strongest problems with alcohol misuse (Bloomfield, Stockwell, Gmel & Rehn, 2003). Some people exhibit moral obligations towards alcohol in these countries. The countries in central Europe are expected to be somewhat in the middle in their attitudes towards alcohol, alcohol is not part of everyday life, however it is not as tucked away as it is in the dry countries. Other researchers have divided the countries in Nordic and

Mediterranean Europe or to the preferred beverage type (i.e. wine, beer or distilled liquor) (Allemani, Hope, Byrne & Room, 2002).

Time series are analyzed by using the CAPM, the three-factor model of Fama and French (Fama & French, 1993) and the four-factor model Jegadeesh and Titman (1995). Evidence is found to support the assumption that differences in cultures matter for performance of alcohol stocks, stocks of alcohol producing companies in central Europe have annually higher abnormal risk-adjusted returns of 0.36% compared to ordinary stocks within the same region. This finding is not found in countries who have a morally neutral attitude towards alcohol.

(5)

4 that the sin stock premium is primarily driven by the cultural environment of investors, i.e., a sin stock premium is not found in countries who morally disapprove alcohol.

The rest of the paper is organized as follows. Section 2 describes the findings of other researchers and the necessary background for the preparation of the hypotheses. Section 3 presents the data and methodology, and Section 4 reports the results for the impact of cultural preferences for alcohol on stock returns. The conclusions and a discussion is provided in Section 5.

2. Literature review

In order to properly describe all the relevant variables related to this subject, first the reasons why investors screen will be explained, afterwards the consequences of the screening will be shown. Furthermore the home bias, the reason why it would matter for investors and companies where the company is located. Finally the drinking cultures within Europe will be explained and this will be linked to financial theories, from which the hypotheses will be derived.

2.1. Screening reasons

During the last decade, Socially Responsible Investing (SRI), also known as sustainable investing or ethical investing, became a wider applied form of investment (Renneboog et al., 2008). SRI encompasses a variety of investments analyses, such as negative screening (also known as exclusionary screening), positive screening, full ESG integration (explicitly including ESG risks and opportunities into all processes of investment analysis and management) and thematic investing (US SIF Foundation, 2012). The focus of this paper lies on negative screening, since sin stocks, and thus stocks of alcohol producing companies, are being avoided by investors using negative screening principle of SRI.

(6)

5 to them (Statman, Fisher & Anginer, 2008). Another explanation could be found from financial rationality; investors might think sin companies will be less profitable in the future, due to a variety of reasons, such as litigation risks (by paying fines to offset the negative impact the company has on society) (Salaber, 2007) or additional excise taxes. A country or a supranational institution such as the European Union might impose these taxes in the future, when firms are not able to pass these additional taxes to customers, they will be charged for it. Furthermore investors might believe a sinful company might experience diminishing returns from consumers, due to an increase in the public’s awareness (Bénabou & Tirole, 2010) and increasing concerns from activists and nongovernmental organizations (Chatterji, Levine & Toffel, 2009). 1

Chatterji et al. (2009) characterized the motivations of investors into financial, deontological, consequentialist and expressive, where financial investors expect sin companies will have inferior financial performance, due to a variety of reasons as mentioned above, deontological investors believe it is unethical to profit from companies that put externalities on society, consequentialist investors want to increase the cost of capital for irresponsible firms and reduce it for virtuous companies (i.e. consequentialist investors use their investments to punish or reward firms) and expressive investors want to express their personal identity in the investments they make, so they do not want to invest in companies they perceive as vicious.

2.2. Outcomes of screening

Hong and Kacperczyk (2009) show that moral constraints will drive investors towards certain behavior, this is validated by Durand et al. (2013), who state investors herd toward virtuous investment behavior consistent with their social norms. They propose that the herding is not only driven by individual investors putting pressure on professional money managers, but that professional money managers themselves appear to be influenced by their personal values as well. This implicates that fund managers do not only strive for value maximization, but are affected by social norms. This is validated by Hong and Kostovetsky (2012) by demonstrating that the political preference of fund managers affect investment decisions. Merton (1987) has developed a model in which is shown that asset segmentation can have an important effect on asset prices. He states that when many investors neglect a certain type of asset this will create limited risk-sharing

(7)

6 opportunities which leads to a significant downward pressure on stock prices. This is further acknowledged by Salaber (2013).

Fabozzi, Ma and Oliphant (2008) find that sin portfolios outperform the market on a total return and risk-adjusted basis; as a possible explanation they state there is a cost in applying social standards, which is not incurred when including sin stocks. In other words, when there are less firms to choose from when creating the optimal portfolio (due to the screening constraint), this portfolio should render a lower return. The same conclusion has been found by Hong and Kacperczyk (2009) who argue that sin stocks outperform the market as they are less likely to be held by norm-constrained institutions.

2.3. Drinking culture

Hong and Kacperczyk (2009) focus their research on the market of the United States, although they argue that social norms might also influence investors outside the US. When they extend their study to countries outside America, to Canada, France, Germany, Italy, The Netherlands, Spain, Switzerland and the United Kingdom, they group the stocks from these countries into one portfolio. While it can be argued that these countries are similar to the United States and to each other on some points, there are probably just as many reasons to state the opposite (Österberg & Karlsson, 2002).

For example there are the differences in culture. For this paper, of considerable importance will be how inhabitants of a country perceive alcohol as sinful. The European Union, by means of The European Comparative Alcohol Study (ECAS) and later the Gender, Alcohol and culture: An International Study (or GENACIS project), puts in a considerable amount of effort to measure the differences within its member countries with respect to attitudes towards alcohol. It is important for European policy makers to be aware of the differences for policy implications. A number of important topics to the European Union are: wine and viticulture as agricultural activity, a common wine policy, barriers to trade in the form of discriminatory taxation, state imposed standards and monopolies, ensuring free movement of television broadcasts, the harmonization of excise duties and preventive alcohol policies (Österberg & Karlsson, 2002).

(8)

7 and English speaking countries: Denmark, Finland, Iceland, Norway and Sweden, Ireland and the U.K. (Allemani et al., 2006; Levine, 1992; Mäkelä, Gmel, Grittner, Kuendig, Kuntsche, Bloomfield & Room, 2006). The wet-dry distinction was initially based on the average amount of alcohol consumed in a country and the prevalence of drinking, but is now based on more characteristics, such as “tradition, religion, social position, income, occupation, gender, region, and a host of other factors” (Rahav, Wilsnack, Bloomsfield, Gmel & Kuntsche, 2006). Table 1 provides an overview of the differences between the wet and the dry drinking culture. Wet drinking cultures are characterized by a high consumption per capita, low or moderate alcohol consumption quantity at a time, a low proportion of abstainers, frequent situations were drinking is common and widespread mechanisms of informal social control of drinking, which means that governmental regulation is not strict and people correct each other in case of alcohol abuse (Rahav et al., 2006; Room & Mitchell, 1972). People from wet countries thus drink frequently, however consume only moderate amounts of alcohol. Opposed to the wet drinking countries are the dry drinking countries, which can be characterized by a low consumption per capita, a high proportion of abstainers, but infrequent very heavy or binge drinking and restricted access (alcohol can often only be bought in governmental owned stores) (Rahav et al., 2006; Room and Mitchell, 1972). When drinking occurs in dry countries, it is more likely to result in intoxication or even alcohol poisoning (Bloomfield et al. 2003). The countries of central Europe are expected to lie somewhere in between these drinking cultures (Mäkelä et al., 2006), which implicates that these countries do have not as many restrictions on alcohol as in the Nordic countries, but alcohol is not part of their culture as in the Mediterranean countries. As the name suggest, they are average in their beliefs on alcohol.

Table 1: Differences between wet and dry drinking cultures

Wet Dry

Alternative names Mediterranean

Southern Europe Moderate Integrated Non-temperance Nordic Northern Europe Immoderate Ambivalent Temperance

Consumption per capita High Low

Quantity consumption at a time Low or moderate High

Proportion of abstainers Low High

Occasions of consumption Frequent Infrequent

(9)

8

Temperance tradition Weak Strong

Type of alcohol Wine Beer or distilled liquor

Availability Widely Restricted

Adapted from Cagney (2006)

Wet countries have a tradition with alcohol, which is assumed to go as far back as the Romans and in these societies alcohol is integrated into daily life and activities (Bloomfield et al. 2003). This becomes visible in the way people from the wet countries favor drinking places, they tend to desire highly visible drinking-places in which the customers as well as the facilities are clearly displayed. Mediterranean countries have a warm climate in which permanent outdoor tables are common, but even during colder days, they favor open spaces and large windows in which a glassed-in pavement is not uncommon. (SIRC, 1998). On the other hand, the dry societies, which are characterized by the (colder) Northern Europe, prefer their drinking places to be more insular, and contained. Bar’s and pubs have walls made of concrete and the establishment has often private partitions, which ensures that what the customers do inside will be kept to themselves. Drinking is kept more secretive than in the wet countries.

Furthermore, dry cultures have a history of strong temperance movements, likely caused by the dominant protestant religion in these countries, which generated a ‘cultural ambivalence’ towards alcohol. On the one hand citizens of dry countries drink less frequently and socially disapprove alcohol, on the other hand when alcohol is consumed, it is consumed at greater quantities at the time, which may lead to alcohol related problems. This is supported by Leifman (2002) who found that intoxication-oriented drinking was most common in Finland, Norway, Sweden and the United Kingdom. These countries face the strongest problems with alcohol misuse, even though there is a significantly less per capita pure alcohol consumed than in wet countries (Levine, 1992). Dry cultures have thus the most concerns over alcohol related problems and their attitudes are the most restrictive. The central European countries are somewhere in between, for example Levine (1992) classifies The Netherlands as a ‘non-temperance’ culture, however he states that it differs from the wet countries in that it has experienced strong Temperance movements, albeit that these movements were of little influence and short-lived (but still present today to some degree).

(10)

9

2.4. Home bias

In order to make assumptions about the impact of cultural values of investors residing in either wet, central or dry countries, it is important to show where these investors make their investments. Following Baker and Nofsinger (2002), investors trade in familiar securities. This is also known as the familiarity bias (Cao, Han & Hirschleifer, 2011; Grinblatt & Keloharju, 2001); investors tend to buy the stocks of companies that have a local or regional business presence, also known as local bias (Huberman, 2001) or suffer from a home bias (French & Poterba, 1991). This means that investors buy a

disproportionate number of securities from their own country or countries located within the same region, although it is widely accepted that substantial gains can be acquired through international diversification. Investors suffer from home bias because companies from their own country or region are more familiar to them than foreign companies (Baker and Nofsinger, 2002). Furthermore currency risks (Othmani, Saanoun, Garali and Arab, 2014); poor governance (Kho, Stulz and Warnock, 2009); or a variety of other reasons (the home bias remains a puzzle [Coval & Moskowitz, 1999]) influence people to invest domestically.

Table 2 gives an illustrative overview of the equity home bias given in percentages researched by Kho et al. (2009); Schoenmaker and Bosch (2009) and Sørensen, Wu, Yosha and Zhu (2007), who use different techniques to measure the level of home bias in a given country. While there are differences in the percentages of equity home biases, due to different data sets2_{, what can easily be recognized is that all} researchers found significant home biases in all countries, with a home bias as high as around 90% in Greece to a moderate bias in the Netherlands (around 40%). Othmani et al. (2014) found there was a steady decline in the bias for most countries during 2001 – 2008, however this was followed by an increase during the years 2008 – 2012.

2_{Kho et al. (2007) base their findings on the data of surveys held by the U.S. treasury department on foreign} holdings of U.S. investors. Schoenmaker et al. (2009) use the IMF Coordinated Portfolio Investment Survey (CPIS) as data source. Sørensen et al. (2007) use data from Lane and Milesi-Ferritti (2006) and Standard & Poor’s Global Stock Markets Factbook. The researchers calculate equity home bias (EHB) in a similar way: Kho et al. (2009) calculate bias as “one minus the ratio of the weights of the country in U.S. equity portfolios and in the world market portfolio”; Schoenmaker et al.(2009) use: 𝐸𝐻𝐵𝑖= 1 − _{𝐹𝑜𝑟𝑒𝑖𝑔𝑛 𝑒𝑞𝑢𝑖𝑡𝑦 𝑡𝑜 𝑇𝑜𝑡𝑎𝑙 𝑀𝑎𝑟𝑘𝑒𝑡}𝐹𝑜𝑟𝑒𝑖𝑔𝑛 𝐸𝑞𝑢𝑖𝑡𝑦𝑖

𝑖,

(11)

10 Therefore it is assumed that most investors invest in their own country and expected stocks returns can be different across countries, as risk premiums depend primarily on country-specific factors (Salaber, 2013).

Table 2: Equity home bias

Equity home bias in %

Kho et al.a _{Schoenmaker et al.}b _Sørensen_{et al.}𝑐 _Average

Austria 76 68 39 61 Belgium 82* 69 50 60 Denmark 78 74 63 72 Finland 57 75 65 66 France 79 79 71 76 Germany 76 77 54 69 Greece 87 97 96 93 Netherlands 49 43 37 43 Portugal 83 85 68 79 Spain 84 93 86 88 Sweden 76 73 58 69 United Kingdom 62 80 68 70

* In their research Belgium is coupled with Luxembourg

A) Calculated for 2004. Kho et al. use the surveys of the U.S. Treasury department of foreign portfolio holdings of U.S. securities

B) Calculated for 2004. Schoenmaker et al. use the IMF Coordinated Portfolio Investment Survey (CPIS) as source. C) Calculated for 2003. Sørensen et al. use a mix of sources.

2.5. Drinking culture and finance

(12)

11 This paper examines whether investors located in dry countries require a higher return on sin stocks than on other stocks. Merton (1987) found that “an increase in the relative size of the firm's investor base will reduce the firm's cost of capital and increase the market value of the firm.” If investors located in dry countries indeed exhibit risk aversion towards companies involved with the production of alcohol, the opposite will happen and these investors will not hold stocks of these sin companies, which means that the investor base of sin companies will be significantly reduced. This in turn will increase the firm’s cost of capital and reduces the market value of the firm, pushing down the price of sin stocks and increasing their expected returns. This is further acknowledged by Heinkel, Kraus & Zechner (2001), who show that ethical investing can influence a firm’s cost of capital, and so affect investment. While some investors neglect the sin stocks and significantly underweight them in their portfolio, others are willing to arbitrage these undervalued stocks, but expect a higher return for the limited risk sharing (Salaber, 2013). Economic logic suggest that these price effects should be temporary, because of arbitrage effects. However Fama and French (2007) argue that when assets are seen as consumption goods, which means that investors have a (dis)taste for certain stocks (in this case some investors refuse to hold stocks of companies involved with alcohol), than this price effect can be long-lasting. As a result (arbitraging) investors residing in dry countries should require higher expected returns on stocks from alcohol producing companies compared to other stocks. This is not the case in wet countries, were investors do not possess sin aversion, since alcohol is not considered as a sin as it is described earlier. Investors in sin stocks in these countries are thus not required to make an abnormal return. The hypotheses this paper wants to test are therefore:

H1. In dry countries, stocks of alcohol producing companies have significant higher risk-adjusted returns than common stocks located in these countries.

H2. In dry countries, stocks of alcohol producing companies have significant higher risk-adjusted returns than similar companies in wet countries.

Investors from central Europe are not expected to be as sin averse as investors from dry countries, however they are expected to be sin averse nonetheless. Since investors in central Europe are, as explained before, less sin averse than the dry countries, however more sin averse than the wet countries. This leads to the following additional hypotheses:

H3: In central countries, stocks of alcohol producing companies have significant higher risk-adjusted returns than common stocks located in these countries.

(13)

12 H5. In central European countries, stocks of alcohol producing companies have significant higher

risk-adjusted returns than similar companies in wet countries.

Since central Europe lies in between the dry and the wet countries, the differences between the alphas of the additional hypotheses are not to be expected as strong as in hypothesis 2.

3. Data and empirical method

This study focuses on the impact of a drinking culture on expected stock returns in a sample of 12 European countries over the period 1990–2014. We believe that the European market is the best setting for our analysis as the European drinking cultures are well described. Furthermore, the Europeans as a whole have the highest alcohol consumption rate and are the largest producers of alcohol (World Health Organization, 2004).

3.1. Drinking culture and other country level data

Table 3 shows which countries are located in which region. As previously suggested, the countries are split into three regions; dry, which consists of Denmark, Finland, Iceland, Ireland, Norway and Sweden; Central Europe, which consists of Austria, Belgium, The Netherlands and Germany; and wet which consists of France, Greece, Italy, Portugal and Spain.

Table 3. Regions of drinking cultures

Dry countries Wet countries Central countries

Denmark.𝐚 France Austria

Finland Greece Belgium

Iceland* Italy* Germany

Ireland* Portugal The Netherlands

Norway* Spain

Sweden

United Kingdom.𝐛

*not used in the sample because it has none or only one publicly listed stock involved with alcohol

A) Ramstedt (2001) argues Denmark has other cultural values than dry countries, therefore the hypotheses are tested with and without the inclusion of Denmark in the sample.

(14)

13

3.2. Financial data

All common stocks available in Thomson DataStream for the sample countries are included, except financial stocks. Monthly financial data on the return index, market capitalization and ratio price-to-book value are downloaded from Thomson DataStream from January 1990 to June 2014. The analysis is based on euro rate of returns, in order to lessen the effect of currency changes. The European risk-free rate is downloaded from Kenneth French’s website3_{. Stocks of companies involved with the distribution and} production of alcohol are selected according to the DataStream classification into “brewers” and “distillers & vintners”. In order to include the dead stocks in the analysis, the “Datastream industrial classification number and name extended description” is used of each company in the Beverages category in DataStream to check whether it is indeed an alcohol stock, since the beverages category includes soft-drinks as well. The category distillers & vintners only include companies involved with alcohol. Dead stocks are included in the download, so that the stocks that merged, defaulted or delisted are not excluded from the sample. This ensures that the data is relatively free of survivorship bias (Salaber, 2013). All companies who miss data on the location, sector and type are deleted, hereby following the methodology of Ince and Porter (2006). Furthermore companies for which data is not available for the total return index are discarded from the sample, as well as outliers (i.e., returns lower than -300% or higher than 300%) and the month June 2014 as the month is not yet finished. After cleaning the data a total of 120 stocks of alcohol producing companies and 8931 non-financial common stocks are left. The low number of alcohol producing stocks is because most sin companies are privately held and is line with the research performed by Hong and Kacperczyk (2006) and Salaber (2013).

For the analysis seven portfolios are constructed: three portfolios of sin stocks (listed either in wet, central or dry countries) and three portfolios with all the available common stocks within the respective region. Furthermore a market portfolio is constructed, with all the common stocks from all the countries in the sample included. All portfolios are value-weighted each month and mutually exclusive. The descriptive

(15)

14 statistics are shown in Table 4. For each group of countries, this table indicates the number of companies involved with the distribution and production of alcohol, the total number of companies in the country sample, the average market value and the average price-to-book ratio. It furthermore shows the percentage of alcohol companies with respect to the total sample. In the early years of the sample not all the stocks were listed, thereby decreasing the number of stocks in the sample, which is in line with Salaber (2013).

Table 4. Country statistics

Number of firms Alcohol / Average Market value Average PTB Alcohol Market Market Alcohol Market Alcohol Market

Austria 5 152 3.29% 182.98 480.41 1.56 2.26 Belgium 4 256 1.56% 8190.78 654.29 1.34 1.46 Denmark 8 267 3.00% 1103.7 582.08 1.76 2.42 Finland 2 195 1.03% 102.54 882.05 2.05 2.28 France 28 1337 2.09% 689.67 1243.03 3.09 2.74 Germany 29 1056 2.75% 71.2 1068.28 5.48 2.31 Greece 3 409 0.73% 27.76 181.98 2.74 1.86 The Netherlands 4 710 0.56% 7770.57 2524.4 3.09 2.45 Portugal 2 91 2.2% 239.63 508.43 1.61 3.13 Spain 9 167 5.39% 243.47 2270.62 1.89 2.34 Sweden 4 807 0.5% 245.75 540.82 1.29 2.81 United Kingdom 22 3484 0.63% 7833.92 1008.01 3.58 2.67 Total 120 8931 1.34%

Notes: Market value is the share price multiplied by the number of ordinary shares in issue and is given in euro millions. PTB is the price-to-book ratio.

3.3. Methodology

(16)

_𝑝𝑡

(2)

4. Empirical results

(18)

17 Table 5. Regression results for equally-weighted portfolios of dry, central and wet countries

This table presents the coefficients of the regression results for the CAPM, the three-factor model of Fama and French (1993) and the four-factor model of Jegadeesh and Titman (1995) for the equally-weighted portfolio of stocks of alcohol producing and non-financial companies. The results are based on monthly returns. The R² shows the adjusted R². In the parenthesis the standard errors are shown. The equations for the models are the following:

(1): 𝑹𝒑𝒕− 𝑹𝒇𝒕𝒈 = 𝜶𝒑+ 𝜷𝒑(𝑹𝒎𝒕𝒈 − 𝑹𝒇𝒕𝒈) + 𝜺𝒑𝒕

(2): 𝑹𝒑𝒕− 𝑹𝒇𝒕𝒈 = 𝜶𝒑+ 𝜷𝒑(𝑹𝒎𝒕𝒈 − 𝑹𝒇𝒕𝒈) + 𝒔𝒑𝑺𝑴𝑩𝒕+ 𝒉𝒑𝑯𝑴𝑳𝒕+ 𝜺𝒑𝒕

(3): 𝑹𝒑𝒕− 𝑹𝒇𝒕𝒈 = 𝜶𝒑+ 𝜷𝒑(𝑹𝒎𝒕𝒈 − 𝑹𝒈𝒇𝒕) + 𝒔𝒑𝑺𝑴𝑩𝒕+ 𝒉𝒑𝑯𝑴𝑳𝒕+ 𝒎𝒑𝑴𝑶𝑴𝒕+ 𝜺𝒑𝒕

Dry Countries Central Countries Wet Countries

Alpha Beta SMB HML WML R² Alpha Beta SMB HML WML R² Alpha Beta SMB HML WML R²

A lcoh o l 1): .0023* (.0012) .8982*** (.0663) .393 .0014** (.0006) .3981*** (.0316) .3593 -.0000 (.0013) .6820*** (.0687) .2573 2): .0017 (.0012) .9299*** (.0708) -.064 (.0567) 1556*** (.0493) .4135 .0012** (.0006) .3888*** (.0343) .0191 (.0275) .0211 (.0239) .3575 .0003 (.0013) .7132*** (.0746) -.0637 (.0597) -.058 (.0519) .2578 3): .0006 (0012) .939*** (.0697) -.0783 (..0558) .2004*** (.0503) .0981*** (.0295) ..4338 0.0012** (.0006) .3889*** (.0344) .0189 (.0276) .0219 (.0249) .0016 (.0146) .3552 .0000 (.0013) .7148*** (.0748) -.0663 (.0599) -.0499 (.054) .0176 (.0317) .256 No n -fi na n ci al 1): .0004 (.0004) 1.0731*** (.0236) .8805 -.0011*** (.0003) .8335*** (.0196) .8647 .0007 (.0006) 1.0933*** (.0276) .8475 2): .0004 (.0004) 1.0587*** (.0256) .0294 (.0247) -.0162 (.0178( .8805 -.0011*** (.0214) .8365*** (.0214) -.0061 (.0171) .0074 (.015) .864 .0007 (.0005) 1.1048*** (.03) -.0233 (.024) .0088 (.0209) .8471 3): .0004 (.0004) 1.0587*** (.0257) .0294 (.0206) -.0106 (.0185) -.0003 (.0109) .8801 -.001*** (.0003) .8354*** (.0214) -.0044 (.0172) .002 (.0154) -.0115 (.0091) .8642 .0006 (.0005) 1.1059*** (.0301) -.025 (.0241) .0142 (.0217) .0118 (.0127) .847

(19)

18 Table 6. Regression results for dry countries with U.K. and Denmark excluded

This table presents the coefficients of the regression results for the CAPM, the three-factor model of Fama and French (1993) and the four-factor model of Jegadeesh and Titman (1995) for the equally-weighted portfolio of stocks of alcohol producing and non-financial companies. The results are based on monthly returns. The R² shows the adjusted R². In the parenthesis the standard errors are shown. The equations for the models are the following:

(1): 𝑅𝑝𝑡− 𝑅𝑓𝑡𝑔 = 𝛼𝑝+ 𝛽𝑝(𝑅𝑚𝑡𝑔 − 𝑅𝑓𝑡𝑔) + 𝜀𝑝𝑡

(2): 𝑅𝑝𝑡− 𝑅𝑓𝑡𝑔 = 𝛼𝑝+ 𝛽𝑝(𝑅𝑚𝑡𝑔 − 𝑅𝑓𝑡𝑔) + 𝑠𝑝𝑆𝑀𝐵𝑡+ ℎ𝑝𝐻𝑀𝐿𝑡+ 𝜀𝑝𝑡

(3): 𝑅𝑝𝑡− 𝑅𝑓𝑡𝑔 = 𝛼𝑝+ 𝛽𝑝(𝑅𝑚𝑡𝑔 − 𝑅𝑓𝑡𝑔) + 𝑠𝑝𝑆𝑀𝐵𝑡+ ℎ𝑝𝐻𝑀𝐿𝑡+ 𝑚𝑝𝑀𝑂𝑀𝑡+ 𝜀𝑝𝑡

Dry Countries (U.K. excluded) Dry Countries (Denmark excluded)

Alpha Beta SMB HML WML R² Alpha Beta SMB HML WML R²

A lcoh o l 1): .0037** (.0016) 1.1004*** (.0881) .3544 .0022 (.0015) .8923*** (.0838) .2875 2): .003* (.0016) 1.1378*** (.0947) -.0757 (.0758) .1752*** (.066) .3691 .0015 (.0016) .9349*** (.0899) -.0864 (.0719) .1689*** (.0626) .303 3): .0017 (.0017) 1.1484*** (.0936) -.0925 (.075) .2278*** (.0676) .1148*** .0396 39.41 .0002 (.0016) .946*** (.0885) -.104 (.071) .2239*** (.064) .1201*** (.0375) .3254 No n -fi na n ci al 1): .0011** (.0005) 1.1093*** (.029) .8383 .0005 (.0005) 1.1557*** (.029) .8493 2): .0012** (.0005) 1.0947*** (.0315) .0296 (.0252) -.0259 .0219 .8389 .0007 (.0005) 1.1405*** (.0313) .031 (.025) -.0458 (.0218) .8516 3): .0012** (.0006) 1.0944*** (.0315) .0301 (.0253) -.0274 (.0228) -.0034 (.0134) .8384 .0007 (.0006) 1.1403*** (.0313) .0312 (.0226) -.0467 (.0226) -.002 (.0133) .851

*significant at the 10% level, **Significant at 5% level, ***significant at 1% level

Regarding hypothesis two, the alpha coefficients of alcohol stocks are higher than the alpha coefficients of similar stocks in the wet countries. Furthermore the alpha coefficients of alcohol producing companies in the wet countries are nearly zero for the CAPM and the four-factor model and slightly above zero for the three-factor model and all insignificant which is consistent with the theory that investors in wet countries are morally neutral towards alcohol. However, due to the insignificant alphas in the three- and four-factor models (and the significant HML and WML factors) in the dry countries, which implies that the alphas are not significantly different from zero, it is not possible to accept hypothesis two.

(20)

19 central European countries. For the CAPM, the alcohol stocks have a 0.03% higher abnormal excess returns per month (0.36% annually), for the three-factor model a monthly abnormal return of 0.01% (0.12% annually) and for the four-factor model an abnormal return of 0.02% (0.24% annually). This is in line with the theory that investors are moderately sin averse (i.e. there are no “big” differences expected, since only a few investors neglect the alcohol stocks from their portfolio). Furthermore, the SMB, HML and WML factors are not statistically significant, which means the CAPM is the best model to explain the returns of the alcohol portfolio. Therefore it is possible to accept hypothesis three; stocks of alcohol producing companies in central European countries have significant higher risk-adjusted returns than ordinary stocks located in the same countries.

While the alpha coefficients are higher in the dry countries with respect to the central countries for the CAPM and the three-factor model, it is impossible to accept hypothesis four, for the same reasons as explained above, namely the alphas are not significantly different from zero. Hypothesis five however can be accepted, as the alpha coefficients of central European alcohol producing company stocks are significantly positive (at the 5% level) and higher than the (insignificant) alphas of the wet countries, in all the applied models. For the CAPM the alpha coefficient is 0.14% higher per month in the central companies than in the wet companies (1.68% annually), .09% for the three-factor (1.08% annually) and 0.12% for the four-factor model (1.44% annually).

Finally the coefficients of the excess market return over the risk-free rate of the common stocks are higher (and closer to one) than the portfolios of alcohol stocks in all regions, which indicates that alcohol stock are less sensitive to market risk exposure.

5. Conclusion and discussion

(21)

20 in wet countries cannot make an abnormal excess return on alcohol producing companies, which is in line with the theory that citizens from these countries do not exhibit sin aversion against alcohol.

A limitation to this paper is the low number of stocks of alcohol producing companies. Using a different data set, could offer more validation. Furthermore, the borders of drinking cultures are set rather artificially as culture is not bound to country borders. Other typologies could be invented to avoid the disadvantages of the nation approach. Also, in one country there could be several minority cultures, which are neglected by using only the dominant culture. In addition, culture is hard to measure, a significant level of research over many years by many researchers has been put to describe the drinking cultures as proficient as possible, however the opinions of people are subjective and are subject to change (although slowly). Finally, alcohol companies might be internationally held. Though investors suffer from a home bias, this does not necessarily implicate that companies are only held be domestic investors. Alcohol producing companies are often large international conglomerates producing for all the countries in the sample. Additional research could provide more validation whether alcohol producing companies are indeed domestically held.

References

Allamani, A., Beck, F., Bergmark, K. H., Csemy, L., Eisenbach-Stangl, I., Elekes, Z […] & Mendoza, M. R. (2006). Gender, culture and alcohol problems: a multi-national study. Oxford Journals.

Allamani, A., Hope, A., Byrne, S., & Room, R. and the ECAS research team (2002) Lessons from the ECAS study: comments and policy implications. Consumption, drinking patterns, consequences and policy responses in 15 European countries, 187-220.

Allamani, A., Voller, F., Kubicka, L., & Bloomfield, K. (2000). Drinking cultures and the position of women in nine European countries. Substance Abuse, 21(4), 231-247.

Baker, H. K., & Nofsinger, J. R. (2002). Psychological biases of investors. Financial Services Review, 11(2), 97-116.

Bénabou, R., & Tirole, J. (2010). Individual and corporate social responsibility. Economica, 77(305), 1-19. Bloomfield, K., Stockwell, T., Gmel, G., & Rehn, N. (2003). International comparisons of alcohol consumption.

Alcohol Research and Health, 27(1), 95-109.

Cagney. P. (2006). A Healthy drinking culture: A search and review of international and New Zealand literature (Final Report), Research New Zealand #3451

(22)

21 Chatterji, A. K., Levine, D. I., & Toffel, M. W. (2009). How well do social ratings actually measure corporate social responsibility? Journal of Economics & Management Strategy, 18(1), 125-169.

Coval, J. D., & Moskowitz, T. J. (1999). Home bias at home: Local equity preference in domestic portfolios. The

Journal of Finance, 54(6), 2045-2073.

Durand, R. B., Koh, S., & Tan, P. L. (2013). The price of sin in the Pacific-Basin. Pacific-Basin Finance Journal,

21(1), 899-913.

Fabozzi, F. J., Ma, K. C., & Oliphant, B. J. (2008). Sin stock returns, Journal of Portfolio Management 35, 82–94. Fama, E. F., & French, K. R. (1998). Value versus growth: The international evidence. The Journal of Finance,

53(6), 1975-1999.

French, K. R., & Poterba, J. M. (1991). Investor diversification and international equity markets (No. w3609).

National Bureau of Economic Research.

Grinblatt, M., & Keloharju, M. (2001). How distance, language, and culture influence stockholdings and trades. The Journal of Finance, 56(3), 1053-1073.

Heinkel, R., Kraus, A., & Zechner, J. (2001). The effect of green investment on corporate behavior. Journal of

financial and quantitative analysis, 36(04), 431-449.

Hong, H. G., Kacperczyk, M. T. (2009). The Price of Sin: The Effects of Social Norms on Markets. Journal of

Financial Economics 93(1), 15–36.

Hong, H., & Kostovetsky, L. (2012). Red and blue investing: Values and finance. Journal of Financial

Economics, 103(1), 1-19.

Huberman, G. (2001). Familiarity breeds investment. Review of financial Studies, 14(3), 659-680.

Ince, O. S., & Porter, R. B. (2006). Individual equity return data from Thomson Datastream: Handle with care!.

Journal of Financial Research, 29(4), 463-479.

Jegadeesh, N., & Titman, S. (1995). Overreaction, delayed reaction, and contrarian profits. Review of Financial

Studies, 8(4), 973-993.

Kho, B. C., Stulz, R. M., & Warnock, F. E. (2009). Financial globalization, governance, and the evolution of the home bias. Journal of Accounting Research, 47(2), 597-635.

Kumar, A., & Page, J. (2011). Deviations from norms and informed trading. .Journal of Financial and

Quantitative Analysis (JFQA), Forthcoming.

(23)

22 Leifman, H. (2001). Trends in population drinking. T. Norström (red.). Alcohol in postwar Europe:

consumption, drinking patterns, consequences and policy responses in 15 European countries

Levine, H. G. (1992). “Temperance Cultures: Alcohol as a Problem in Nordic and English-Speaking Cultures” in Malcom Lader, Griffith Edwards, and D. Colin Drummon (ed) The Nature of Alcohol and Drug-Related

Problems. New York: Oxford University Press, 1993, pp.16-36

Othmani, S., Saanoun, I. B., Garali, W., & Arab, M. B. (2014). Determinants of Home Bias Puzzle in European Countries. International Review of Management and Business Research, 3(1)

Mäkelä, P., Gmel, G., Grittner, U., Kuendig, H., Kuntsche, S., Bloomfield, K., & Room, R. (2006). Drinking patterns and their gender differences in Europe. Alcohol and Alcoholism, 41(suppl 1), i8-i18.

Merton, R. C. (1987). A simple model of capital market equilibrium with incomplete information. The journal

of finance, 42(3), 483-510.

Norström, T. (2001). Per capita alcohol consumption and all‐cause mortality in 14 European countries.

Addiction, 96(1s1), 113-128.

Österberg, E., & Karlsson, T. (2002). Alcohol policies in EU member states and Norway. A collection of country reports.

Rahav, G., Wilsnack, R., Bloomfield, K., Gmel, G., & Kuntsche, S. (2006). The influence of societal level factors on men's and women's alcohol consumption and alcohol problems. Alcohol and Alcoholism, 41(suppl 1), i47-i55.

Ramstedt, M. (2001) Per capita alcohol consumption and liver cirrhosis mortality in 14 European countries,

Addiction, 96, Supplement 1, S19-S34

Ramstedt, M. (2002). Alcohol-related mortality in 15 European countries in the postwar period. European

Journal of Population/Revue européenne de Démographie, 18(4), 307-323.

Renneboog, L., Ter Horst, J., & Zhang, C. (2008). Socially responsible investments: Institutional aspects, performance, and investor behavior. Journal of Banking & Finance, 32(9), 1723-1742.

Room, R., & Mäkelä, K. (2000). Typologies of the cultural position of drinking. Journal of Studies on Alcohol

and Drugs, 61(3), 475.

Salaber, J. (2007). The determinants of sin stock returns: Evidence on the European market.

Salaber, J. (2013). Religion and returns in Europe. European Journal of Political Economy, 32, 149-160. Schoenmaker, D., & Bosch, T. (2008). Is the home bias in equities and bonds declining in Europe?. Investment

(24)

23 SIRC (1998). Social and Cultural Aspects of Drinking, A report to the European Commission. Available at: http://www.sirc.org/publik/social_drinking.pdf

Skog, O. J. (1985a). The Collectivity of Drinking Cultures: A Theory of the Distribution of Alcohol Consumption.

British journal of addiction, 80(1), 83-99.

Skog, O. J. (1985b). The wetness of drinking cultures: a key variable in epidemiology of alcoholic liver cirrhosis. Acta Medica Scandinavica, 218(S703), 157-184

Skog, O. J. (2001). Alcohol consumption and mortality rates from traffic accidents, accidental falls, and other accidents in 14 European countries. Addiction, 96(1s1), 49-58.

Sørensen, B. E., Wu, Y. T., Yosha, O., & Zhu, Y. (2007). Home bias and international risk sharing: Twin puzzles separated at birth. Journal of International Money and Finance, 26(4), 587-605.

Statman, M., Fisher, K. L., & Anginer, D. (2008). Affect in a behavioral asset-pricing model. Financial Analysts

Journal, 20-29.

Drinking Culture and Stock Returns in Europe