1
Bachelor Thesis Finance & Organization
-University of Amsterdam-
Abstract
This thesis studies the performances of sin stocks compared with regular stocks. The entire United States stock market is divided into 49 portfolios. Four of these portfolios are considered to be sinful because these products are supposed to be harmful to individuals, communities or the environment. The performances of all portfolios are measured.
This study found evidence that the sinful portfolios are valued differently than the regular portfolios.
This thesis investigated additionally the differences in relative performances before September 2008 and after September 2008.
This thesis found no significant evidence for the different treatment of the sin stocks before September 2008 and after September 2008.
Name: Niels Leijten
Studentnumber: 6088481 Supervisor: Timotej Homar
Date: 29/06/2015
2
Statement of Originality
This document is written by Student, Niels Leijten, who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Introduction
Sin companies are companies involved, or associated, with immoral or unethical activities. Unwritten ethical norms force managers to eliminate these sin firms out of their portfolios. There is no specific framework to justify whether an industry or company could be
considered as sinful or not. The degree of sin of an industry is changing over time, because our norms and values about sinfulness are subject to change. Peer effects from civil society are important indicators for the economic behavior and the market outcomes nowadays (Kubler, 2001).
Socially responsible investing (SRI) is not a new phenomenon. In the American Colonies, the Quakers avoided already investments that would benefit the slave trade. In the 1980s, when the anti-Apartheid campaigns in South Africa took place, SRI developed at lightning speed. Social investors and institutions avoided all companies associated with South Africa as a protest against racial discrimination. (Social Investment Forum, 2005)
In the past decades, the theory of socially responsible investments has become an
important fundament of the investment decision for investors. Approximately 10 percent of the total assets under management involve socially responsible investing (Social Investment Forum, 2005). Socially responsible investment (SRI) means that non-financial criteria, such as social, governmental, environmental and ethical criteria are involved into the investment process (Eurosif, 2010). In light with this growing perception of SRI, investors are becoming less willing to invest in sin stocks although the returns are relatively high.
Corporate social screening (CSS) is the practice of evaluating investment portfolios or mutual funds based on social, ethical and environmental criteria. This screening process is part of the social responsible investing strategy. Social investors avoid investing in
companies whose products and business practices are harmful to individuals, communities, or the environment. Social investors can either screen on positive or negative elements. Positive screening includes the ethnic diversity and sustainability of investments. Negative screening focuses on sinful businesses and immoral activities such as extraordinary working
4
conditions. In either case, screenings narrows the investment area. This means that screening potentially worsens the benefits of investing according to the portfolio theory. The classical portfolio theory says that investors should base their investment decisions only on risk measurements and return rates.
The Social Investment Forum investigated the social screens of mutual funds in 2005. They found that ‘tobacco’ is the most commonly used social screen. This means that tobacco products are the most eliminated industry when constructing a portfolio. Alcohol is the second largest dismissed industry in a portfolio. The ‘defense’ and ‘soda & candy’
companies are less often eliminated from a portfolio. The latter could be considered as less sinful since there are evidently less objections to them.
Table 1: Social Screens Mutual Funds
Industry Assets 1)Tobacco $159 2)Alcohol $135 3)Gambling $41 4)Defense $34 5)Candy&Soda $2
(Social investment forum, 2005)
In order to clarify the issue of this thesis, it investigates the performance of sinful industries. This leads to the research question of this study: Are companies, involved or associated with immoral or unethical activities valued differently than the overall market?
The null hypothesis states that companies enrolled into unethical or immoral activities are not valued differently than regular companies. This means that there is no way to
distinguish sin stocks from the market, based on their returns. The alternative hypothesis is in this case the fact that a different valuation approach between sin stocks and regular stocks is used. This should be visible in the performances of the stocks.
The additional regression investigates the differences between the valuation of the
5
the worldwide credit crisis started in September 2008 when Lehman Brothers Bank
collapsed. The additional hypothesis is that there is no difference in the valuation of the sin stocks. This could be argued by the lack of substitutes and the addictiveness of sinful products.
To answer the questions, US stock market data is gathered from the Kenneth R. French library. The excess returns and abnormal returns of sinful portfolios are compared with the regular portfolios. Two variables were added to the model to control for market
capitalization and book-to-market value.
The entire US stock market is divided into 49 portfolios. Five of them showed a significant abnormal return. The tobacco portfolio is the only portfolio that outperforms the market on a 1% significance level. The average weighted returns of the tobacco portfolio outperform the market with an abnormal return of 1.27% according to the regression test. Defense does outperform the market as well with an abnormal return of 0.86% on a 5% significance level. The ‘alcohol’ and ‘candy & soda’ portfolios are not significant at all.
The results of the additional regressions are different. Before September 2008, six portfolios are significant on a 5% level but these are all regular portfolios. After September 2008, five different portfolios are significantly outperforming the market. These portfolios are also regular portfolios. This suggests that no sinful portfolios significantly outperforming the market when the time span is shorter. This is striking because the tobacco industry and the defense industry do have a significant abnormal return when the entire period is taken into account.
This thesis will consist of several sections. The following part continues with an overview of the existing relevant literature. Then, the methods and data used will be explained and described. The next part analyses the findings and finally, the conclusion of thesis research will be stated in the last section.
6
Literature
A sinful company is a ranging concept without fixed criteria. Whether a company is sinful depends partly on personal opinions. There exists no specific framework whether an industry is sinful or not. Previous literature focusses mainly on the tobacco, alcohol and gaming industry when it comes to a research about sin stocks. These industries do have in common that their products are addictive and undesirable for social society. (Hong and Kacperczyp, 2009)
The most cited article within this subject is written by Hong and Kacperczyp (2009). Hong and Kacperczyp (2009) investigated the performance of 184 sin stocks in the gaming, tobacco and alcohol industry in the American market between 1926 and 2006. They claim that these ‘classic sin stocks’ outperform the market In the United States because of the avoidance of the big fund managers. This means that it would be financial beneficial to invest in those stocks. They concluded that the sin stocks, partly due to the avoidance of large investment companies, the market outperformed by 3.5%.
Other researchers investigated different markets but they found, in contradiction to the US market, no significant evidence for an outperformance. This indicates that the matter of outperformance of sin stocks is not totally clear yet.
Kim and Venkatachalam (2006) investigated the fact that sin stocks may have greater information risk because of poor financial reporting quality. They concluded that the sin firms’ financial reporting quality is even better than their control group. They indicated that the sin stocks became financially attractive due to the high quality financial reporting. They also suggest that investors are just willing to pay an extra financial ‘premium’ in order to comply with societal terms and norms. Furthermore, American sin stocks are also
investigated during economic recessions. Salaber (2009) found that sin portfolios also during a recession outperform the market.
7
There has also been research about the performance of sin stocks in the Asian market. Visaltanachoti et al. (2009) did research on the performance of sin stocks In China and Hong Kong. They found significant evidence on a 1% level for an excess return of 3.3%. The sin stocks in Europe also outperform the market according to the study of Salabar (2009). She found a significant excess return of 4% per year under the CAPM model.
Lobe and Walkshausl (2011) investigated the sin stock performances in 51 countries all over the world. They divided the countries in specific sub regions. However, in contrast with others, they did not find any statistically significant outperformance of their sin stock sample. The performances of the sin stocks are not yet so clear.
Data and methods
This chapter will provide an inside into the methods used to determine the performances of the stocks of the sinful industries.
This study focusses on the stocks of the entire United States stock market. The main reason to focus on the US stock market is the relatively easy access to the well maintained datasets compared with other countries. Furthermore, the US stock market provides the biggest sample of sin stocks so there is enough data available to investigate. In some European countries are for example not enough listed sinful companies to run a significant regression. There are for instance more tobacco companies situated in the United States, than in Europe. (Lobe and Walkhäusl, 2011).
The time span for this research is from January 2000 to December 2014. I have chosen for this period because, as far as I was able to ascertain, no other research had covered this specific period yet. On the other hand, social responsibility became an important factor in the investments decisions in the 21th century. The number of investment funds
incorporating environmental, social and ethical criteria in their investments decisions grew since the start of this millennium from 168 funds to 925 in 2014 (US SIF, 2014). January 2000 was also chosen as starting point because this study will investigate the performance of sin stocks within the 21th century. After investigating the overall performances during 15 years,
8
I will run two additional regressions. One will be from January 2000 until August 2008 and the other one will be from September 2008 until December 2014. This distinction is made to determine whether there is a difference in the performances between those periods. The first period indicates the performances before the credit crisis and the second period indicates the performances after the start of the crisis.
This thesis used the Kenneth R. French website to gather relevant financial data. This website is linked with the Center for Research in Security Prices (CRSP) and updated at least once a year. They publish convenient overviews of returns of several portfolios within in US stock market. The data used for this thesis divided the entire stock market into 49
portfolios. These portfolios include all companies listed on the NYSE, AMEX and NASDAQ. All companies are classified by their SIC codes and they belong to one of the portfolios. A SIC code is a Standard Industrial Classification Code and it is used to identify certain
establishments. For instance, the 6th portfolio is called ´Smoke´. All companies with a SIC code between 2100 and 2199 are added to this portfolio. These firms are all considered to be tobacco related firms. These 49 different portfolios differ off course in size. The ´Personal Services´ portfolio includes for instance all funeral companies, barber shops, electronica repair shops etc.
This thesis investigates the monthly weighted average returns of the 49 portfolios. The time span of the study is 15 years so this means that 180 months are investigated for each portfolio. This number of observations is enough to do reliable research.
Four portfolios will be labeled as sin; the ‘tobacco’, ‘alcohol’, ‘defense’ and ‘candy & soda’ portfolio. The existing research investigates often the tobacco, alcohol and gambling
industry. These industries are generally speaking called the ‘classic sin industries’. This study deviates from the establishments because the potential sin portfolios where chosen on certain criteria. First, the social responsible investments criteria were used to determine whether an industry could be named as sin or not. Secondly, when an industry is sin, it should be classified as an own separate portfolio according to the division made by Kenneth R. French. The third portfolio made by Kenneth R. French is called ‘candy & soda’ and it includes all companies with SIC code 2064, 2065, 2066, 2077, 2068, 2086, 2087, 2096 and
9
2097. ‘Gambling’ and ‘pornography’ are subdivided into two overarching portfolios. These portfolios are called ‘entertainment’ and ‘personal services’. This is the main reason why the gambling and pornography stocks are excluded in this thesis.
The weighted average returns of every month of every portfolio is calculated by Kenneth R. French by multiplying the weights and the returns of all companies within the portfolio and divide it by the number of companies. The Fama and French (1993) control variables are also gathered from the Kenneth R. French data library.
The ‘size’ variable (SMB) is included to overcome ‘small firm effects’. Small firms usually tend to outperform larger firms. The SMB variable is calculated by subtracting the weighted average return of the large companies by the weighted average return of the small ones. A company is defined as a large company when their market capitalization is higher than the median market capitalization of the whole market. Similarly, when the market capitalization of a firm is lower than the median market capitalization, the firm is defined as a small firm. The second control variable in the model is the value variable (HML). Companies with a relative high market ratio tend to outperform the companies with lower market ratios according to Fama & French. This is called a value premium. The book-to-market ratio will be determined by dividing the book value of equity in the previous year by the market value of equity at the end of the previous year. A firm’s book-to-market ratio is defined as high when the ratio is higher than the 70th percentile. A ratio is considered as a low book-to-market ratio when the ratio is below the 30th percentile. The value variable is calculated by subtracting the companies with a high book-to-market ratio from the companies with a low book-to-market ratio.
When all relevant data is collected and schematized in an excel sheet, the regression model will come to light. 49 separate regressions are made to test the hypothesis whether the performances of the sin stocks is valued differently. The performances of the portfolios are all tested by estimating the following model:
10
Where r(pft,j) is the monthly return of a portfolio, rft is the monthly risk free rate. This rate is
subtracted from the monthly portfolio returns. The answers of these subtractions are the dependent variables in the regressions. SMBt is the variable ‘size’ and the HMLt is the ‘value’
variable. These are the independent variables. The beta1, beta2 and beta3 are included to
test the effect of the variables on the portfolio returns. If the portfolio correlates more with large firms, it would be visible into the beta2. The value of beta3 shows whether the
portfolio correlates more with high book-to-market value firms or not. The assumption made for these predictions would be that all other factors are kept constant. The β0 shows
the predicted outperformance of the portfolio compared with the market. The initial hypothesis will be rejected when the β0 of a sinful portfoliosignificantly differs from 0. This
would mean that the tested portfolio is valued differently than the market. Analysis
This section will discuss the findings of the regression tests. The results of the individual portfolios are presented in the following figures and tables.
The following table gives on overview of the significant portfolios.
The weighted average returns of all portfolios are regressed on the Fama-French factors . For each industry are the α, β1, β2 and β3 schematized in the appendix. The ‘α’ corresponds
to the abnormal return of the industry. This abnormal return is compared with the entire market and it is controlled for the SMB ratio and the HML ratio. These variables are denoted by β1, β2 and β3.
Only 5 out of 49 portfolios reflect an abnormal return with a significance level of at least 5%. The other 44 tested portfolios are not significant in this regression model.
The abnormal return of the tobacco industry is 1.27% and that is also the only portfolio with a significance level of 1%.
The sin portfolios ‘alcohol’ and ‘candy & soda’ do not have a significant outperformance at all. The tobacco portfolio seems to be the eye-catching portfolio within the sin market.
Although only five portfolios have a significant abnormal return, almost all betas are significant. This means that the Fama French regression does work very well.
11
Table 2: Significant Portfolios (2000-2014)
Significance at 1%***; 5%**; 10%*
Figure 1 shows the monthly average excess returns of all portfolios between 2000 and 2014. The excess returns are calculated each month by subtracting the risk free rate from the portfolio returns.
The average excess returns of the ‘tobacco’ portfolio, the ‘defense’ portfolio and the ‘candy & soda’ portfolio are respectively 1.62%, 1.35% and 1.06%. It is striking that these unethical portfolios perform relatively well although they are considered as unethical. The fourth sinful portfolio is ‘alcohol’. The average excess return of this portfolio is 0.60%. Even though this return is considerably less than the other sinful portfolios, it is still higher than the average excess return of the market. The average excess return of the entire market is 0.33% per month. An investor would have gained the highest return (1.64%) by investing into ‘shipbuilding & railroad equipment’. Investing in the ‘communication’ portfolio yielded the lowest average monthly excess return (0.05%).
Significantie Portfolio Alpha Beta1 Beta2 Beta3
P ≤ 0,01 Smoke 1,2711*** 0,6302*** -0,4169*** 0,4114*** (2,74) (6,06) (-3,09) (2,97) P ≤ 0,05 Guns 0,8635** 0,5257*** -0,3336*** 0,6059*** (2,08) (5,65) (-2,76) (4,88) Clths 0,6836** 1,1041*** -0,2788*** 0,2409** (2,01) (14,47) (-2,81) (2,37) Meals 0,5645** 0,7943*** -0,2513*** 0,182** (2,19) (13,75) (-3,35) (2,36) Food 0,5084** 0,4943*** -0,3135*** 0,2678*** (2,14) (9,27) (-4,53) (3,76) P ≤ 0,10 Boxes 0,5676* 1,0619*** -0,1308 0,023192 (1,8) (15,06) (-1,43) (0,25) Util 0,5215* 0,5322*** -0,3014*** 0,2680*** (1,88) (8,59) (-3,75) (3,24) Drugs 0,4284* 0,5949*** -0,2133*** 0,0439 (1,69) (10,47) (-2,89) (0,58) Ships 0,8462* 1,2245*** -0,2781** 0,6828*** (1,86) (12,01) (-2,10) (5,01)
12
Figure 1
Figure 2 illustrates the betas of all portfolios. These betas reflect the risk of investing in a portfolio relative to the entire market. The beta of the entire market is 1. The steel portfolio has the highest beta of 1.76. In case the prices of the entire market would rise with 10%, the prices within the steel portfolio would rise with 17.,6%. But the other way around counts as well. When the prices in the market would drop with 10%, the steel portfolio would drop 17,6%. The lower the beta, the less sensitive a portfolio is to market fluctuations.
Figure 2 shows that the selected sin portfolios all have a relatively low beta compared with the other portfolios. The ‘candy & soda’ portfolio is the one with the highest beta (0.61). ‘Tobacco’ has a beta of 0.52, the ‘defense’ portfolio has a beta of 0.42 and the beta of alcohol is only 0.34. An increase of 1 euro in the market price leads only to an increase of 33 cents in the ‘Alcohol portfolio’.
‘Gold’ is the only portfolio that is less sensitive to the market performance with a beta of 0.30. These portfolios do not exhibit a large sensitivity to economic market conditions. This suggests that investing in the sin portfolios could be a relatively safe investment.
sin sin sin sin market 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 Ships Sm o ke C o al G un s C lt hs Tx tl s So da M ine s A gr ic Fun Hlt h B o xe s M eal s M ac h A er o Tr an s R lE st C ns tr Oil C hem s Ut il Fo o d B ldM t R ub br M edEq Whl sl P ap e r P e rS v B ee r La bEq D rug s Ins ur To ys Fin El cE q R ta il Fab P r H sh ld H ar dw A ut o s G o ld B an ks St ee l B us Sv O th er B o o ks So ft w C hi ps Te lc m M ar ke t
13
Figure 2
Figure 3 shows the ‘alpha’ values of the portfolios. The alpha is the intercept of the
regression with the Y-axis. The alpha determines how much the realized return of a portfolio differs from the expected return. This corresponds to the abnormal return of a portfolio and it reflects the return that cannot be explained by the systematic market risk and the Fama-French factors.
The abnormal returns of the sinful portfolios are high compared with the other portfolios. According to this data, the tobacco industry (1.27) and the defense industry (0.86) are the ones with the highest abnormal return of all industries. The ‘candy & soda’ portfolio has an abnormal return of 0.69 and the ‘alcohol’ portfolio has a return of 0.47.
sin sin sin sin market 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 St ee l C hi ps H ar dw Fun R lE st A ut o s M ac h La bEq Txtl s So ft w M ine s El cE q Fi n C o al C ns tr Fab P r B ldM t Shi ps C he m s B us Sv B oxes B o o ks C lt hs R ub br Te lc m A er o O th er B an ks To ys Ins ur Tr an s P ap e r W hl sl R ta il P e rS v Oil A gr ic M eal s M edEq Soda Hlt h Drug s Sm o ke H sh ld Ut il G un s Fo o d B ee r G o ld M ar ke t Portfolio Beta's (2000-2014)
14
Figure 3
Table 3: Significant Portfolios (2000-2008)
Significance at 1%***; 5%**; 10%* sin sin sin sin -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Sm o ke G un s Shi ps C lt hs So da A gr ic B o xe s M eal s Ut il Fo o d M ine s B ee r Oil D rug s C o al Tr an s A er o C he m s R ub br M ac h H lt h M edEq Hsh ld W hl sl P e rS v Fun Rtai l La bEq P ape r H ar dw So ft w Tx tl s G o ld El cE q Ins ur C ns tr B ldM t To ys Fin B us Sv Te lc m Fab P r R lE st C hi ps B an ks O th er A ut o s St ee l B o o ks Portfolio Alpha's (2000-2014)
Significantie Portfolio Alpha Beta1 Beta2 Beta3
P ≤ 0,05 Agric 1,2465** 0,4955*** 0,2201** -0,0019 (2,33) (4,13) (2,07) (-0,01) Books -0,7348** 0,8280*** -0,1144 0,2795** (-2,47) (8,89) (-1,23) (2,49) Mach 0,9489** 1,2668*** 0,0874 -0,0508 (2,51) (11,9) (0,76) (-0,36) Mines 1,5096** 1,0238*** -0,0994 0,1645 (2,13) (6,87) (-0,49) (0,69) Oil 1,119** 0,7504*** -0,3772*** 0,1990 (2,19) (5,44) (-3,48) (1,35) Boxes 0,9716** 1,0431*** -0,2413 0,0345 (2,16) (7,44) (-1,39) (0,17) P ≤ 0,10 Smoke 1,3026* 0,6883*** -0,4163** 0,5996*** (1,8) (2,81) (-2,02) (2,88) Chems 0,6072* 0,9226*** -0,3474*** 0,1447 (1,80) (9,94) (-3,41) (1,06) Coal 2,5614* 0,8040** 0,4516 0,4309 (1,91) (2,01) (1,11) (0,82)
15
The above table shows the abnormal returns of the significant portfolios from January 2000 until September 2008. During this period are six portfolios significant on a 5% level. Almost all betas are significant so this means that the Fama-French model is doing well.
In contrast with table 2 are also negative abnormal returns visible.
The most striking thing is that almost all portfolios are different than the once who are stated in table 2. During this period are no abnormal returns of the sinful portfolios
measured with a sufficient significant level. The tobacco industry is the only sinful portfolio with an abnormal return of 1.3026 on a 10% significance level.
Table 4: Significant Portfolios (2008-2014)
Significance at 1%***; 5%**; 10%*
Significantie Portfolio Alpha Beta1 Beta2 Beta3
P ≤ 0,01 Steel -1,7085*** 1,7004*** 0,5831* -0,0871 (-2,95) (9,86) (1,86) (-0,3) Coal -3,1446*** 1,9307*** -0,0273 -0,5365 (-2,66) (6,23) (-0,05) (-0,93) P ≤ 0,05 Food 0,6304** 0,5715*** -0,1552 0,0106 (2,2) (7,93) (-1,12) (0,08) Drugs 0,6941** 0,7466*** -0,2492 -0,2981** (2,39) (10,69) (-1,5) (-2,4) Meals 0,6976** 0,7143*** 0,1845 0,0006 (2,47) (13,04) (1,17) (0,01) P ≤ 0,10 Mines -1,4682* 1,8926*** -0,2215 0,6368** (-1,84) (9,53) (-0,6) (-2) Cnstr -0,8235* 1,2297*** 0,6830*** 0,1807 (-1,95) (11,55) (3,67) (1,04) Soda 1,0983* 0,7110*** 0,2125 0,3225 (1,79) (4,18) (0,65) (1,34) Beer 0,6984* 0,6057*** -0,5064*** -0,0443 (1,86) (5,09) (-3,16) (-0,25)
16
Table 4 reflects the abnormal returns of the significant portfolios after September 2008. Again, five portfolios are significant with at least 5%. The impact of the credit crisis seems to be visible because this table shows many more negative abnormal returns. During this period are the ‘alcohol’ portfolio and ‘candy & soda’ portfolio significant at 10%.
According to the data, the ‘tobacco’ industry and the ‘defense’ industry were not significant at all within this period.
17
Conclusion
This final section will summarize and conclude this thesis.
The results of this study are partly aligned with the results of some previous studies about the performance of unethical stocks in the United States.
The sinful portfolios seem to outperform the regular stocks in the United States between 2000 and 2014. Even after controlling for the model risk factors. (SMB and HML)
However, the results showed a distinction between the sinful portfolios. Tobacco has the highest weighted average excess return and the highest alpha but not the highest beta. This means that it has the highest return, but not the most risk.
When the entire period is divided into two shorter periods is no outperformance visible anymore. The ‘tobacco’ industry has for instance an significant abnormal return of 1.27 in 15 years but it is not significant at all when the period is shortened. This suggests that the performances of sin stocks do not depend on the sinfulness of a portfolio but simply on a stable demand.
A limitation of this thesis would be that it was not possible to investigate the performances of the ‘classic’ sin stocks. This thesis follows the classifications of the Kenneth R. French database. ‘Alcohol’ and ‘tobacco’ were classified as separete portolios, but ‘gambling’ was not. This industry was subornated to the ‘entertainment’ portfolio.
A suggestion for further research would be to investigate the degree of sin within the sinful portfolio. Would it for instance be possible to address each portfolio to a certain level of sin? Previous research often gathers all sin stock together to investigate them against the market. This study suggests that the tobacco industry is the core portfolio with respect to the outperformance. This study indicates that the outperformance of the sin stocks will not be significant anymore when the tobacco industry will be excluded.
18
References
Fabozzi, F. and Ma, K.C. and Oliphant, J.B., 2008. Sin stock returns. The Journal of Portfolio Management.
Fama, E.F. and French, K.R., 1993. Common risk factors in the returns on stocks and bonds. Journal of Financial Economics 33.1: 3-56.
Hong, H. and Kacperczyk, M., 2009. The price of sin: The effects of social norms on markets. Journal of Financial Economics 93.1: 15-36.
Kim, I. and Venkatachalam, M., 2008. Are Sin Stocks Paying the Price for their Accounting Sins? Working Paper, Duke University.
Kubler, D., 2001. On the regulation of Social Norms. Journal of Law, Economics, and Organization.
Lobe, S. and Walkshäusl, C., 2011. Vice vs. Virtue Investing Around the World. Working paper, University of Leicester.
Salaber, J., 2007. The determinants of sin stock returns: Evidence on the European market." Finance International Meeting AFFI- EUROFIDAI Paper.
Social Investment Forum, 2006. Report on Socially Responsible Investing Trends in the United States, 10-year review.
US SIF Foundation, the Forum for Sustainable and Responsible Investment. 2012. Report on sustainable and responsible investing trends in the United States.
Visaltanachoti, N., Zheng Q., Liping, Z., 2009. The performance of sin stocks in China. Working paper, Massey University.
19
Portfolio Alpha Beta1 Beta2 Beta3 Portfolio Alpha Beta1 Beta2 Beta3 1 Agric 1,2465** 0,4955*** 0,2201** -0,0019 26 Guns 1,0677 0,1677** -0,3314** 0,4804 (2,33) (4,13) (2,07) (-0,01) (1,66) (1,00) (-2,09) (2,18) 2 Food 0,2352 0,4575*** -0,3409*** 0,3406*** 27 Gold 0,3900 0,2261 0,1603 0,4241 (0,65) (4,77) (-3,25) (2,91) (0,41) (0,95) (0,87) (1,61) 3 Soda 0,4196 0,4343** -0,1156 0,1777 28 Mines 1,5096** 1,0238*** -0,0994 0,1645 (0,62) (2,16) (-0,61) (0,7) (2,13) (6,87) (-0,49) (0,69) 4 Beer 0,1207 0,2870** -0,2435 0,1558 29 Coal 2,5614* 0,8040** 0,4516 0,4309 (0,30) (2,13) (-1,47) (0,95) (1,91) (2,01) (1,11) (0,82) 5 Smoke 1,3026* 0,6883*** -0,4163** 0,5996*** 30 Oil 1,119** 0,7504*** -0,3772*** 0,1990 (1,8) (2,81) (-2,02) (2,88) (2,19) (5,44) (-3,48) (1,35) 6 Toys -0,1225 0,8278*** 0,1833 0,2004 31 Util 0,4771 0,5673*** -0,3351** 0,3985** (-0,24) (6,49) (1,1) (0,94) (1,13) (4,28) (-2,24) (2,37) 7 Fun 0,5994 1,2704*** -0,0096 0,0070 32 Telcm -0,4397 1,1451*** -0,2240** -0,0878 (1,27) (8,35) (-0,07) (0,04) (-0,99) (11,16) (-2,42) (-0,63) 8 Books -0,7348** 0,8280*** -0,1144 0,2795** 33 PerSv 0,4638 0,7079*** -0,0651 0,1696 (-2,47) (8,89) (-1,23) (2,49) (0,98) (6,32) (-0,36) (0,93) 9 Hshld 0,3971 0,2847* -0,0771 -0,1211 34 BusSv -0,2072 1,0571*** 0,0581 0,0406 (1,18) (1,73) (-0,32) (-0,62) (-0,87) (11,99) (0,66) (0,42) 10 Clths 0,7123 1,0589*** -0,3974** 0,2386 35 Hardw 0,2838 1,7624*** 0,5038** -0,4452* (1,65) (6,89) (-1,94) (1,1) (0,55) (12,55) (2,47) (-1,79) 11 Hlth 0,0220 0,4477** -0,3460 0,6635** 36 Softw 0,2951 1,4407*** 0,3540*** -0,6811*** (0,04) (2,63) (-1,13) (2,42) (0,81) (13,74) (2,67) (-4,57) 12 MedEq 0,4065 0,4723*** 0,0767 0,1540* 37 Chips -0,0223 1,7658*** 0,6038*** -0,4638** (1,35) (5,79) (1,02) (1,76) (-0,05) (11,84) (2,64) (-2,03) 13 Drugs -0,0511 0,5382*** -0,1996 0,1502 38 LabEq 0,2629 1,4956*** 0,5909*** -0,1811 (-0,14) (4,54) (-1,32) (0,9) (0,65) (11,95) (4,41) (-1,12) 14 Chems 0,6072* 0,9226*** -0,3474*** 0,1447 39 Paper 0,0273 0,7843*** -0,3615** 0,2723* (1,80) (9,94) (-3,41) (1,06) (0,08) (7,82) (-2,46) (1,75) 15 Rubbr -0,2211 0,7089*** 0,3434** 0,2179 40 Boxes 0,9716** 1,0431*** -0,2413 0,0345 (-0,51) (5,67) (2,44) (1,23) (2,16) (7,44) (-1,39) (0,17) 16 Txtls 0,2608 0,8321*** -0,0447 0,2029 41 Trans 0,3939 0,7945*** -0,1843 0,1947 (0,43) (4,89) (-0,22) (0,83) (1,04) (7,08) (-1,51) (1,51) 17 BldMt 0,1841 0,8472*** -0,1201 0,2630 42 Whlsl 0,1089 0,7619*** -0,0272 0,2269* (0,49) (8,61) (-0,75) (1,63) (0,33) (7,85) (-0,28) (1,76) 18 Cnstr 0,5681 1,1679*** 0,0806 0,4678** 43 Rtail -0,0451 0,9075*** -0,1663 0,1525 (1,05) (7,87) (0,43) (2,25) (-0,12) (7,69) (-1,4) (1,16) 19 Steel 0,5647 1,7549*** 0,1870 0,0946 44 Meals 0,3548 0,8709*** -0,3399** 0,2952* (1,09) (13,58) (1,64) (0,64) (0,92) (7,85) (-2,06) (1,76) 20 FabPr 0,3398 0,8967*** 0,2422 0,0141 45 Banks -0,0187 0,9346*** -0,4860*** 0,3146* (0,51) (4,69) (1,39) (0,07) (-0,05) (7,8) (-3,62) (1,95) 21 Mach 0,9489** 1,2668*** 0,0874 -0,0508 46 Insur 0,0214 0,8862*** -0,6027*** 0,4783*** (2,51) (11,9) (0,76) (-0,36) (0,06) (9,75) (-4,44) (3,53) 22 ElcEq 0,4076 1,2803*** -0,1491 0,0357 47 RlEst -0,0583 0,6562*** 0,1430 0,3188** (1,19) (12,79) (-1,35) (0,32) (-0,13) (5,95) (1,15) (2,05) 23 Autos -0,5566 1,3473*** -0,262603 0,3219 48 Fin 0,1972 1,6884*** 0,0245 0,2369** (-0,86) (7,47) (-1,18) (1,19) (0,61) (19,11) (0,31) (1,90) 24 Aero 0,3479 1,1486*** -0,4828** 0,45048** 49 Other -0,0242 0,7562*** -0,2068 -0,0134 (0,61) (6,23) (-2,53) (2,02) (-0,05) (4,49) (-1,42) (-0,08) 25 Ships 0,8710 0,8119*** -0,4482*** 0,5238** (1,34) (3,96) (-2,74) (2,37) significance at 1%***; 5%**; 10%* Appendix
20
Table 2: 49 Portfolio performances (2000-2008)
Portfolio Alpha Beta1 Beta2 Beta3 Portfolio Alpha Beta1 Beta2 Beta3 1 Agric 0,6361 0,7026*** 0,1569 0,0241 26 Guns 0,8635** 0,5257*** -0,3336*** 0,6059*** (1,46) (7,19) (0,218) (0,18) (2,08) (5,65) (-2,76) (4,88) 2 Food 0,5084** 0,4943*** -0,3135*** 0,2678*** 27 Gold 0,0497 0,2963* 0,0895 0,2691 (2,14) (9,27) (-4,53) (3,76) (0,06) (1,72) (0,4) (1,17) 3 Soda 0,6780 0,6389*** -0,0820 0,2852** 28 Mines 0,4849 1,3311*** -0,1004 0,2331 (1,46) (6,16) (-0,61) (2,06) (0,93) (11,4) (-0,66) (1,5) 4 Beer 0,4663 0,4170*** -0,2966*** 0,1465* 29 Coal 0,4071 1,1612*** 0,4175 0,5835** (1,63) (6,52) (-3,57) (1,72) (0,44) (5,66) (1,57) (2,13) 5 Smoke 1,2711*** 0,6302*** -0,4169*** 0,4114*** 30 Oil 0,4562 0,8625*** -0,3471*** 0,2586** (2,74) (6,06) (-3,09) (2,97) (1,33) (11,21) (-3,48) (2,52) 6 Toys -0,0328 0,956*** 0,2087** 0,263** 31 Util 0,5215* 0,5322*** -0,3014*** 0,2680*** (-0,1) (12,51) (2,1) (2,58) (1,88) (8,59) (-3,75) (3,24) 7 Fun 0,2275 1,4803*** 0,1094 0,1965* 32 Telcm -0,1205 1,0484*** -0,2032*** -0,1525** (0,59) (17,05) (0,97) (1,69) (-0,53) (20,66) (-3,09) (-2,25) 8 Books -0,4478 1,054*** -0,0238 0,38177*** 33 PerSv 0,2297 0,7957*** 0,0891 0,1322 (-1,59) (16,72) (-0,29) (4,54) (0,61) (9,42) (0,81) (1,17) 9 Hshld 0,2521 0,5327*** -0,1424* 0,0903 34 BusSv -0,0678 1,0509*** 0,096** -0,0138 (0,98) (9,25) (-1,91) (1,18) (-0,40) (27,98) (1,97) (-0,28) 10 Clths 0,6836** 1,1041*** -0,2788*** 0,2409** 35 Hardw 0,1761 1,4351*** 0,4306*** -0,5830*** (2,01) (14,47) (-2,81) (2,37) (0,48) (17,45) (4,03) (-5,31) 11 Hlth 0,2941 0,6763*** -0,1974* 0,6282*** 36 Softw 0,1173 1,2303*** 0,3167*** -0,7159*** (0,75) (7,67) (-1,72) (5,33) (0,51) (23,68) (4,70) (-10,32) 12 MedEq 0,2913 0,6838*** 0,1142 0,1980*** 37 Chips -0,1955 1,4329*** 0,5729*** -0,5785*** (1,20) (12,55) (1,62) (2,72) (-0,58) (19,13) (5,89) (-5,79) 13 Drugs 0,4284* 0,5949*** -0,2133*** 0,0439 38 LabEq 0,1891 1,2496*** 0,6029*** -0,3347*** (1,69) (10,47) (-2,89) (0,58) (0,68) (19,99) (7,43) (-4,01) 14 Chems 0,3361 1,1674*** -0,2948*** 0,2357*** 39 Paper 0,1798 0,9483*** -0,2979*** 0,3402*** (1,29) (19,96) (-3,88) (3,02) (0,70) (16,54) (-4,00) (4,45) 15 Rubbr -0,019 0,9632*** 0,3630*** 0,3854*** 40 Boxes 0,5676* 1,0619*** -0,130776 0,0231921 (-0,06) (13,91) (4,04) (4,17) (1,8) (15,06) (-1,43) (0,25) 16 Txtls 0,0866 1,3972*** 0,1047 0,7271*** 41 Trans 0,3873 0,9178*** -0,1319* 0,2693*** (0,17) (12,27) (0,71) (4,78) (1,51) (15,99) (-1,77) (3,51) 17 BldMt -0,02236 1,1912*** 0,0171 0,4961*** 42 Whlsl 0,2439 0,8579*** 0,0271 0,2015*** (-0,07) (16,69) (0,18) (5,21) (1,11) (17,43) (0,42) (3,07) 18 Cnstr 0,0154 1,1928*** 0,19218* 0,4698*** 43 Rtail 0,2272 0,85834*** -0,1204* 0,0582 (0,04) (16,68) (1,82) (4,33) (0,93) (15,75) (-1,7) (0,8) 19 Steel -0,4083 1,7029*** 0,2766** 0,1163 44 Meals 0,5645** 0,7943*** -0,2513*** 0,182** (-1,11) (20,62) (2,58) (1,05) (2,19) (13,75) (-3,35) (2,36) 20 FabPr -0,154 1,1218*** 0,351** 0,1631 45 Banks -0,2951 1,1182*** -0,425*** 0,5972*** (-0,36) (11,7) (2,82) (1,27) (-1,03) (17,51) (-5,13) (7,01) 21 Mach 0,3301 1,3853*** 0,1554** -0,2530 46 Insur 0,0259 1,0598*** -0,5138*** 0,53478*** (1,23) (23,1) (2) (0,32) (0,11) (19,21) (-7,18) (7,26) 22 ElcEq 0,0421 1,3011*** -0,2137 0,0767 47 RlEst -0,1616 1,2367*** 0,2621** 0,6879*** (0,16) (22,53) -0,29 (0,99) (-0,38) (13,02) (2,13) (5,43) 23 Autos -0,3044 1,4803*** (-0,09) 0,33129* 48 Fin -0,0425 1,475*** 0,5266 0,0998 (-0,67) (14,58) -0,65 (2,45) (-0,17) (26,04) (0,72) (1,32) 24 Aero 0,3843 1,1082*** -0,3951*** 0,3550*** 49 Other -0,3027 1,0719*** -0,2511*** 0,3286*** (1,14) (14,69) (-4,04) (3,52) (-0,97) (15,38) (-2,78) (3,53) 25 Ships 0,8462* 1,2245*** -0,2781** 0,68278*** (1,86) (12,01) (-2,10) (5,01) significance at 1%***; 5%**; 10%*
21
Portfolio Alpha Beta1 Beta2 Beta3 Portfolio Alpha Beta1 Beta2 Beta3
1 Agric -0,6650 1,1523*** -0,2479 -0,5993 26 Guns 0,3058 0,9245*** -0,2839 0,1707 (-0,99) (5,2) (-0,67) (-1,57) (0,63) (7,31) (-1,22) (0,73) 2 Food 0,6304** 0,5715*** -0,1553 0,0106 27 Gold -1,2090 0,7497** -0,4126 -0,6187* (2,2) (7,93) (-1,12) (0,08) (-0,94) (2,15) (-0,57) (-1,69) 3 Soda 1,0983* 0,7110*** 0,2125 0,3225 28 Mines -1,4682* 1,8926*** -0,2215 0,6368** (1,79) (4,18) (0,65) (1,34) (-1,84) (9,53) (-0,6) (-2) 4 Beer 0,6984* 0,6057*** -0,5064*** -0,0443 29 Coal -3,1446*** 1,9307*** -0,0273 -0,5365 (1,86) (5,09) (-3,16) (-0,25) (-2,66) (6,23) (-0,05) (-0,93) 5 Smoke 0,6692 0,8129*** -0,5260*** -0,1875 30 Oil -0,5702 1,0467*** -0,2653 -0,0474 (1,57) (7,56) (-2,69) (-1,25) (-1,54) (10,48) (-1,14) (-0,25) 6 Toys 0,1024 1,0178*** 0,4011 0,2378 31 Util 0,2204 0,6215*** -0,1856 -0,1367 (0,22) (9,87) (1,6) (1,36) (0,59) (6,55) (-0,97) (-0,8) 7 Fun -0,0136 1,4896*** 0,7824** 0,2665 32 Telcm 0,3505 0,9131*** -0,0800 -0,0480 (-0,02) (11,49) (2,31) (1,01) (1,3) (12,24) (-0,65) (-0,39) 8 Books -0,0059 1,1068*** 0,5737** 0,3528* 33 PerSv -0,3379 0,8790*** 0,8446*** -0,3301 (-0,01) (10,59) (2,04) (1,87) (-0,58) (5,82) (3,03) (-1,31) 9 Hshld 0,2571 0,6963*** -0,3799*** 0,2724* 34 BusSv -0,0037 1,0547*** 0,2912*** -0,1558*** (0,88) (9,14) (-3,06) (1,75) (-0,02) (14,14) (3,69) (-2,7) 10 Clths 0,6365 1,0624*** 0,3431 0,1177 35 Hardw 0,0934 1,2307*** -0,0722 -0,3842** (1,29) (8,27) (1,55) (0,58) (0,21) (12,32) (-0,33) (-2,21) 11 Hlth 0,2614 0,8726*** 0,6317*** 0,0529 36 Softw 0,0888 1,0357*** 0,0527 -0,4269*** (0,68) (9,43) (3,55) (0,42) (0,35) (18,15) (0,42) (-3,79) 12 MedEq -0,1112 0,9240*** 0,3327* -0,1776 37 Chips -0,2630 1,1612*** 0,2900** -0,3103** (-0,31) (10,00) (1,86) (-0,78) (-0,86) (14,52) (2,16) (-2,14) 13 Drugs 0,6941** 0,7466*** -0,2492 -0,2981** 38 LabEq 0,0052 1,1014*** 0,5518*** -0,3823** (2,39) (10,69) (-1,5) (-2,4) (0,02) (11,66) (3,92) (-2,58) 14 Chems -0,1996 1,4076*** 0,0019 -0,1024 39 Paper 0,3893 1,0079*** 0,1117 0,2727* (-0,57) (15,05) (0,01) (-0,59) (1,25) (16) (0,58) (1,85) 15 Rubbr 0,4206 0,1050*** 0,6093*** 0,51560*** 40 Boxes -0,0735 1,0804*** 0,3733** -0,2425 (1,04) (12,79) (2,71) (2,8) (-0,21) (11,3) (2,29) (-1,4) 16 Txtls 0,6833 1,3998*** 0,1174*** 1,3314*** 41 Trans 0,4395 0,9491*** 0,1907 0,2589* (0,95) (9,49) (2,8) (3,29) (1,52) (10,28) (1,34) (1,76) 17 BldMt -0,0910 1,2896*** 0,8491*** 0,4865** 42 Whlsl 0,2207 0,9581*** 0,3314*** -0,0707 (-0,21) (12,02) (3,66) (2,51) (0,94) (13,74) (3,66) (-0,52) 18 Cnstr -0,8235* 1,2297*** 0,6830*** 0,1807 43 Rtail 0,4464 0,8205*** 0,1153 -0,0836 (-1,95) (11,55) (3,67) (1,04) (1,63) (10,04) (0,79) (-0,6) 19 Steel -1,7085*** 1,7004*** 0,5831* -0,0871 44 Meals 0,6976** 0,7143*** 0,1850 0,0006 (-2,95) (9,86) (1,86) (-0,3) (2,47) (13,04) (1,17) (0,01) 20 FabPr -0,7631 1,2424*** 0,9311*** -0,0042 45 Banks -0,0448 0,9938*** 0,0052 1,1327*** (-1,36) (10,46) (3,32) (-0,02) (-0,12) (8,53) (0,03) (5,21) 21 Mach -0,5572 1,5119*** 0,4530*** -0,2196 46 Insur -0,0398 1,1547*** -0,0025 0,3376** (-1,58) (15,98) (2,77) (-1,39) (-0,12) (12,38) (-0,01) (2,43) 22 ElcEq -0,3474 1,2261*** 0,6061*** 0,0203 47 RlEst -0,0470 1,4931*** 1,1088*** 0,6943* (-0,96) (16,08) (3,93) (0,14) (-0,07) (10,25) (2,83) (1,74) 23 Autos -0,0129 1,4498*** 0,89960** 0,1385 48 Fin -0,4820 0,1370*** 0,0690 -0,0314 (-0,02) (9,45) (2,4) (0,45) (-1,4) (13,85) (0,42) (-0,2) 24 Aero 0,2379 1,09170*** 0,0188 0,0868 49 Other -0,1632 0,1160*** -0,3293** 0,7968*** (0,67) (11,34) (0,12) (0,44) (-0,47) (15,6) (-2,08) (4,6) 25 Ships 0,6955 1,4609*** 0,7274** 0,3005 (1,09) (7,83) (2,06) (0,93) significance at 1%***; 5%**; 10%*
