Benchmarking Sin

by Craig L. Israelsen
Reprinted from Financial Planning Magazine, February, 2006

he Social Investment Forum (www.socialinvest.org) reports that the three most frequently used "screens" utilized by socially responsible fund managers (i.e. socially responsible investing, or SRI) exclude companies involved in the production of alcoholic drinks, tobacco products, and gambling/hotel casinos. This study addresses the following question: What is the performance differential between these three non-SRI industries and all other industries (in the U.S. and non-U.S. equity markets)?

A persistent question surrounding socially responsible investing is whether or not one must sacrifice performance in order to follow ones individual conscience. Fundamental to this question is determining what types of SRI screening filters are needed to satisfy such a social requirement, for that is what determines which companies are off-limits. Understandably, different people/firms define socially responsible investing differently and will focus on different screens as a result. For instance, other SRI screens exclude firms in aerospace and defense industries, firms with poor labor relations, firms using nuclear power to generate electricity, or firms that have a negative environmental impact.

For the sake of this analysis, three industries (alcohol, tobacco, gambling/hotel casinos) will be defined as being clearly non-SRI and their collective performance will be contrasted against the performance of stocks in all other industries. Raw data for this analysis were obtained from the stock module in the year-end releases of Morningstar Principia from 1999 through 2004 (thus minimizing the impact of survivorship bias).

In the figure "Calibrating Sin", the returns of the three major non-SRI industries are compared against the performance of the 30 largest industries in the world (according to number of shares outstanding) as well as four prominent equity benchmark indexes. The performance of each industry was determined by calculating the equal-weighted return of all the firms within that industry. In addition, the share-weighted, one-year return was calculated. Equal-weighting represents a raw average of all the annual returns, whereas share-weighted returns (as well as market cap-weighted returns) are calculated by assigning a proportionally greater weight to the returns of larger firms based on share percentage or market-capitalization percentage. The industries are listed according to the data in the far right column (Sum of Equal-Weighted Return Ranks). In comparison to the 30 largest industries, the non-SRI or "sin" stocks had stellar performance over 1, 3, 5, and 10 year periods ending 12/31/04. For example, the 31 stocks in the Gambling/Hotel Casino industry had a share-weighted, one-year return of 49.01% in 2004, which ranked 1st. The 12 companies in the Tobacco industry had a share-weighted return of 28.87% in 2004 which ranked that industry 4th. The Alcoholic Drinks industry (comprised of 18 firms) had a share-weighted one-year return of 14.47%, which ranked it 17th among this sample of industries. The one-year, share-weighted return of Gambling/Hotel Casinos was significantly higher than the return of the equal-weighted Wilshire 5000 Index and dramatically higher than the returns of the market-cap weighted Wilshire 5000, the Russell 2000, and the MSCI EAFE Index. Share-weighted returns were not calculated for periods longer than one year inasmuch as share totals can change too much over periods greater than one year.

Using equal-weighted performance, the three sin industries also excelled. Over a one-year period (2004), the 88.47% return of Gambling/Hotel Casinos was over 40 percentage points higher than one-year return of 47.96% for Oil/Gas, the next closest industry. On the basis of the sum of the performance rankings over the 1, 3, 5, and 10 year periods (where the lowest sum is the winner), Gambling/Hotel Casinos was the best performing industry. Oil/Gas was 2nd, Tobacco was 3rd, and Alcoholic Drinks were 12th.

The equal-weighted returns of the three sin industries significantly outperformed the three market capitalization-weighted indexes (Wilshire 5000, Russell 2000, and MSCI EAFE) over the 1, 3, and 5 year periods. Over the ten-year period ending 12/31/04, the Alcoholic Drink industry slightly underperformed the U.S. equity indexes while Gambling and Tobacco outperformed each market-cap weighted equity benchmark.

The equal-weighted Wilshire 5000 Index is a stiffer benchmark. Gambling/Hotel Casinos outpaced it in all periods except over the past 10 years. Tobacco outperformed the equal-weighted Wilshire 5000 Index over the one and five year periods, but lagged over the three-year and 10-year period. Alcoholic Drinks lagged the equal- weighted 5000 in all four periods.

On the basis on industry level performance, investing in sin has been consistently and highly profitable in recent years. It is particularly interesting to note the strong performance of each sin industry over the five-year period ending on 12/31/04. During this period, many other industries had negative five-year annualized returns, as well as the market-cap weighted Wilshire 5000 Index and the MSCI EAFE Index. However, the return of Gambling/Hotel Casinos was ranked 1st, Tobacco was ranked 2nd, and Alcoholic Drinks was ranked 9th. Almost looks like industries that created addictions are somewhat recession-proof.

At year-end 1999, there were a total of 74 companies in the three non-SRI industries (27 in alcohol, 15 in gambling, 19 in hotel casinos, and 13 in tobacco), but only 70 had a one-year return. The average equal-weighted, one-year return of those 70 firms was 26.15% (see "Sin Scorecard"). In other words, at year-end 1999, the one-year return of the imaginary "Sin Fund" (an equal-weighted portfolio of all 70 sin stocks) was 26.15%. (Assuming a sinful expense ratio of 2.00%, the one-year return would have been 24.15%).

The average equal-weighted, one-year return as of 12/31/1999 for the remaining 6,933 "non-sin" firms was 43.68 percent. Clearly not all of the remaining 6,933 firms satisfied all of the additional SRI screens, but they did satisfy the three most common screens. The average return of the 6,933 SRI stocks far exceeded the return of the 70 sin stocks in 1999 (see "Seasons of Sin").

Comparing the average return of 70 sin stocks against the average return of 6,933 SRI stocks poses a potential problem of sample size discrepancy. As a solution, the one-year performance of 100 randomly generated portfolios comprised of 70 SRI stocks (the same as the number of sin stocks in 1999) was calculated using a Monte Carlo-like technique. The average return of those 100 SRI portfolios was 45.59%, very close to the one-year return of 43.68% for all 6,933 SRI firms (see Sin Scorecard). For comparison, the return of the equal-weighted Wilshire 5000 Index in 1999 was 38.4%.

As shown in "Sin Scorecard", over the six years from 1999-2004, the average, equal-weighted return for the socially responsible stocks was very similar to the average, equal-weighted return of the 100 randomly generated portfolios (where each year the number of firms in the randomly generated socially responsible portfolios matched the number of sin stocks). It appears that in the context of this study, despite a large discrepancy in sample size, the average return for all SRI (i.e., non-sin) stocks provides valid data for comparison against a much smaller sub-set of sin stocks.

The annual equal-weighted returns of sin stocks, non-sin stocks, the random sample of non-sin stocks, and the Wilshire 5000 Index for five additional years (2000-2004) are also reported in "Sin Scorecard" and shown graphically in "Seasons of Sin". The performance advantage demonstrated by socially responsible stocks in 1999 disappeared in 2000, 2001, and 2002. During that three-year period, the industries of alcohol, tobacco, and gambling had equal-weighted returns well above the average performance of socially responsible firms and the U.S. equity market in general (as measured by the equal-weighted Wilshire 5000 Index). In 2003, socially responsible firms performed, on average, about as well as the Wilshire 5000 Index, whereas the 71 stocks in the three sin industries significantly underperformed. In 2004, the average performance of 61 sin stocks significantly outperformed the average performance of 5,421 socially responsible stocks and the equal-weighted Wilshire 5000 Index. (The decline in number of SRI companies in 2004 in the Morningstar stock database was due to changes in data sources used by Morningstar during 2004).

As a side note, it is interesting to observe that in 2001, the equal-weighted Wilshire 5000 Index had a return of 27.9%. In that same year, the return of the S&P 500 Index was negative 11.9%. Moreover, 47% of all U.S. stocks and 77% of all U.S. equity mutual funds had a negative one-year return in 2001. As virtually all U.S. equity indexes are market-capitalization weighted, the performance of an equally-weighted index is often significantly different.

Some will argue that performance comparisons such as this should be conducted using weighted data. Thus, to generate comparable performance figures to the market capitalization-weighted Wilshire 5000 Index (the more commonly reported version of the Wilshire 5000), the annual returns for all stocks (both sin and non-sin) were weighted according to their proportion of all shares outstanding.

Weighted annual returns are reported in "The Weight of Sin". The absolute levels of performance changed, but the general trends evidenced using equally-weighted performance data persisted. Sin industries (Tobacco, Alcohol, Gambling/Casino Hotels) under-performed in 1999 and 2003, and out-performed in 2000, 2001, 2002 and 2004. The weighted results are graphically presented in "The Wayward Path".

For those who advocate and engage in socially responsible investing, this type of analysis is not particularly relevant inasmuch as the rationale for SRI is not likely grounded in raw performance, but in philosophical and social commitments. This research simply clarifies what the cost might be in shunning certain industries. As has been shown, industries that create addictions are profitable, at least during the time frame of this study. Those who follow a moral mandate in their investing behavior (and presumably in their personal behavior) exhibit a willingness to avoid industries that have generated impressive returns. Such willingness is impressive-particularly if SRI investors and managers have a realization of what they are passing up.

Calibrating Sin

Sin Scorecard

Seasons of Sin

The Weight of Sin

The Wayward Path

____________________________________________________________________________________
Craig L. Israelsen, Ph.D. is an associate professor at Brigham Young University. He teaches family finance in the Department of Home and Family Living. His research interests include mutual fund analysis. He writes monthly for Financial Planning magazine. Learn more about Craig Israelsen at http://familyliving.familylife.byu.edu/faculty/israelsen.htm