Socially responsible investing ("SRI") involves the integration of social and environmental criteria into the investment decision-making process. As such, it attempts to allow investors to match their personal values to their investment decisions, and to consider both their financial needs and the investment's impact on society. SRI has experienced very rapid growth in Canada in the past few years. As of 2000, the total assets under management in Canadian SRI mutual funds were approximately $6 billion. Despite its recent growth, the development of SRI in Canada has lagged that in other areas of the world, especially the U.S.

One of the major issues surrounding SRI is its effect on performance. There is continual debate as to whether SRI outperforms or underperforms "conventional" (non-SRI) investing. The majority of studies on the topic do not indicate any statistically significant differences in the returns from SRI as compared to conventional investing. However, all of these studies were completed for other countries, and none consider SRI in a Canadian context. As such, the purpose of this study is to investigate whether the Canadian situation is any different.

SOCIALLY RESPONSIBLE INVESTING DEFINED¹
SRI incorporates three main strategies that work together to promote socially and environmentally responsible business practices: (i) shareholder advocacy (or shareholder activism, which involves direct communication with corporate management and boards of directors); (ii) community-based investment (for example, investing in community-based financial institutions, co-ops, and local entrepreneurs); and (iii) screening. The application of screens to the investment process is the "backbone" of SRI and the focus of this study. Negative screening involves excluding securities from portfolios because of policies, practices or products that are considered negative or objectionable. Positive (or affirmative) screening is a more proactive and complex mode of screening that involves including securities in a portfolio because of positive characteristics associated with them. Positive screening may be used to include securities of companies that show leadership in product design, employee policies, environmental protection, human rights, or other positive contributions to society. For positive screening, investors often use a best-of-sector approach, in which a company's record is evaluated in relation to that of its industry counterparts.

SRI is important to both individual investors as well as institutional investors. For example, some pension funds in the U.S. prohibit certain types of investments such as tobacco companies. In Canada, some credit unions rely heavily on this investment approach.

The definition of what is "socially responsible" can vary widely. As such, screens can differ significantly from investor to investor or from fund to fund. However, nearly all socially screened mutual funds screen out tobacco, gambling and alcohol securities, as well as those of companies involved in weapons manufacturing. Most funds also refuse to invest in companies with poor records on environmental, human rights, and labour issues, and an increasing number of funds gear their screens to particular religious beliefs (especially with respect to birth control and abortion) and the welfare of animals. Figure 1a outlines U.S. screens, as reported by the Social Investment Forum, while Figure 1b outlines Canadian screens, as reported by the Social Investment Organization. Finally, other funds have screens relating to product impact on the health and safety of consumers, charitable contributions, community involvement, corruption, and ethical business practices. The majority of investment managers use multiple screens.

Although SRI is a global phenomenon, it is by far most developed in the U.S., which is the largest, most liquid, and most diversified market for SRI. As such, the U.S. market has become the "bellwether" for SRI, and discussions on the topic are often focused on this market.

In November 1999, the Social Investment Forum published its most recent Report on Responsible Investing Trends in the United States, which identified $2.16 trillion in managed portfolios utilizing at least one SRI strategy (see Figure 2). This implies that one out of every eight dollars under professional management in the U.S. is involved in socially or environmentally responsible investing, accounting for roughly 13% of the total $16.3 trillion in assets under management. Furthermore, between 1997 and 1999, the growth in assets involved in SRI significantly outpaced the growth in total assets. Screening is the dominant strategy for U.S. SRI investors. The number of screened mutual funds increased to 175 in 1999, up from just 55 in 1995. SRI mutual funds now encompass all asset classes, including equity, balanced, index, and bond funds, among others.

Historically, SRI has had to fight the perception that it may be better for your soul than your bottom line because of its presumed cost (in terms of lower returns). These attitudes have been challenged in a number of academic studies that have found that socially screened portfolios provide competitive returns to investors.

Proponents of SRI claim that it achieves greater returns because of the additional selection criteria imposed. The basic premise is that the companies that will prosper are well-managed companies, making safe and useful products, treating employees fairly, respecting the environment, and making a positive contribution to their community. Furthermore, proponents argue that screening leads to a reduction in potential liabilities (such as those associated with class action lawsuits or environmental clean-ups), which could negatively affect the earnings of companies included as part of a portfolio.

Critics contend that the additional limitations associated with SRI unduly constrain investment managers. They argue that limiting security choice by criteria other than fundamental or technical analysis compromises the performance of SRI-managed investments because it compels managers to reject a large number of potentially good investments. By imposing further constraints on security selection, social screens restrict the feasible investment opportunities set, and can result in dominated portfolios that exhibit lower returns for a given level of risk, or higher risk for a given level of returns. Critics also claim that the performance of SRI-managed investments may be negatively affected by the increased administrative costs of selecting and monitoring securities to ensure compliance with environmental and social guidelines.

Studies have examined SRI performance at both a relatively general (i.e., non-statistical) level and a more rigorous (statistical) level. In general, results have been mixed. The Social Investment Forum (2000)² analyzed the performance of U.S. SRI mutual funds through mid-2000 by looking at data from Morningstar, Lipper, and Wiesenberger, three U.S.-based organizations that evaluate the performance of mutual funds. The key findings of the study were that over 70% of SRI funds (12 out of 17) with $100 million or more in assets received top rankings from either Morningstar or Lipper, or both, and that ten placed in the top quartile of their investment categories based on Wiesenberger's three-year performance records. Israelsen and Sharpe (1999)³ examined if and how SRI mutual funds differ in meaningful ways from comparable non-socially conscious mutual funds by comparing 16 SRI funds to a group of 24 non-SRI funds. They concluded that, when compared to like-sized mutual funds, socially conscious funds have provided comparable returns over the past five years, after adjusting for loads and taxes. In some academic studies, Hamilton, Jo, and Statman (1993)⁴ examined the relative (risk-adjusted) returns of socially responsible portfolios and conventional portfolios. They employed Jensen's alpha to test the investment performance of 17 SRI mutual funds from 1981 to 1990, and reported that SRI mutual funds did not earn statistically significant excess returns and that their performance was not statistically significantly different from the performance of conventional mutual funds. Statman (2000)⁵ compared the returns of socially conscious and conventional funds and found that the socially responsible mutual funds performed better than the conventional funds of equal asset size, although the difference was not statistically significant. Both the socially responsible funds and conventional funds trailed the S&P 500 by a wide margin.

In our study, we examine the Canadian evidence. We test whether there is a statistically significant difference between the performance of SRI-based investing and conventional (non-SRI) investing (as proxied by the benchmark, described below) in the Canadian market.

In assessing the performance of SRI, we examine the performance of Canadian SRI mutual funds. This investment instrument is most appropriate because it represents how a typical Canadian investor can participate in SRI. Of the 24 mutual funds that screen according to social or environmental criteria, only six of the funds are classified as Canadian Equity or Canadian Large Cap Equity in terms of asset class. We focus on these two asset classes for two reasons: (i) SRI intuitively makes more sense in terms of equity investments as opposed to fixed income investments; and (ii) doing so ensures consistency with the benchmark that we have chosen as a proxy for conventional investing, the TSE300 Total Return Index (the "benchmark"). By choosing this benchmark, we are comparing the performance of SRI mutual funds to the market, as opposed to evaluating the performance of SRI mutual funds relative to conventional (non-SRI) mutual funds. We implicitly assume that the average conventional domestic equity mutual fund performance is similar to the overall market performance. If in fact the average fund underperforms the benchmark, then this is a tougher test of outperformance.

The study considers two different time frames, a five-year period (January 1995 - December 1999) and a ten-year period (January 1990 - December 1999). Applying these time constraints in conjunction with the asset class constraints leaves the following funds available to be tested: for the ten-year period, Ethical Growth and Investors Summa; and for the five-year period, Ethical Growth, Investors Summa, Desjardins Environment, and Clean Environment Equity.⁶ In addition, for each time period, we calculate an equally weighted composite to represent the "average" fund (the "composites").7 Monthly total returns (including price appreciation and any distributions) were obtained directly from globefund.com for each of the time periods for each of the funds. Similarly, for each of the time periods, monthly return data for the TSE300 Total Return Index and 91-day Government of Canada T-bills was obtained from the Canadian Financial Markets Research Centre (CFMRC) database.

For each of the time periods, three different performance measures are calculated to measure the performance of the SRI mutual funds (and the composites) relative to the benchmark.

Mean excess returns are measured as RExcess = RFund - RBenchmark, where RFund is the annualized compound return based on monthly returns for the respective SRI mutual fund and RBenchmark is the annualized compound return based on monthly returns for the TSE300 Total Return Index.

Sharpe ratios (for each of the funds and for the benchmark) are measured as Sharpei = (Ri - RRisk-free)/*Return(i) where Ri is the annualized compound return on the fund or the benchmark, RRisk-free is the annualized compound return for 91-day Government of Canada T-bills and *Return (i) is the annualized standard deviation of the monthly returns for the respective SRI mutual fund or the benchmark.

Jensen's alpha (*) is estimated from the regression equation RFund - RRisk-free = * + ß (RBenchmark - RRisk-free). The reported alpha is annualized.

Two of the three measures (the Sharpe ratios and Jensen's alpha) are risk-adjusted return measures. This is important in the context of evaluating the performance of SRI mutual funds, since the degree of risk of SRI investments is central to both proponents' and critics' arguments. Proponents argue that SRI investments are less risky because the underlying companies are "better"; critics argue that SRI investments are more risky because SRI limits diversification and imposes constraints on the investment opportunity set.

Figure 3 presents the results of the performance measurement analysis. The results indicate underperformance of most of the SRI mutual funds relative to the benchmark, but lower risk exposure. However, in all cases, any underperformance is not statistically significant (at the 95% confidence level). The performance exception is the Investors Summa fund, for which all statistics indicate outperformance.

During the period from January 1990 to December 1999, the annualized compound returns for the two Canadian SRI mutual funds were 9.45% for Ethical Growth and 13.26% for Investors Summa, with the composite earning a return of 11.34%. Over the same time period, the annualized compound return for the benchmark was 11.74% and, for 91-day Government of Canada T-bills, was 6.43%. Of the two funds with ten-year histories, Investors Summa outperformed and the Ethical Growth underperformed the benchmark on the basis of mean excess returns and Jensen's alpha, though for both funds the results were not statistically significant. Similar results were obtained for the Sharpe ratio (although, in this case, no comment can be made with respect to statistical significance). As for the composite, the results indicate underperformance on the basis of mean excess returns and outperformance on the basis of the Sharpe ratio and Jensen's alpha, although in all cases the results are not statistically significant. These results reflect the lower standard deviation (relative to the market's standard deviation) as well as the lower beta of these funds.

During the five-year period from January 1995 to December 1999, the annualized compound returns for Canadian SRI mutual funds ranged from 13.53% to 22.23%, with the composite earning a return of 16.73%. Over the same time period, the annualized compound return for the benchmark was 18.62% and, for 91-day Government of Canada T-bills, was 4.87%. For this five-year time period, on the basis of mean excess returns, only one of the four funds (Investors Summa) outperformed the benchmark, though, once again, none of the results were statistically significant. This same fund was also the only one to show outperformance in terms of the Sharpe ratio. In terms of Jensen's alpha, two of the four funds (Investors Summa and Clean Environment Equity) indicate outperformance, although in only one case was the result statistically significant - the alpha for Investors Summa, with a t-statistic of 2.20. We examined the strong overall performance for Investors Summa and found that it is attributable to consistent performance over the 1990 to 1999 period. Surprisingly, the fund was the largest among the SRI funds. As for the composite, the results indicate underperformance on the basis of mean excess returns, very slight outperformance on the basis of Jensen's alpha, and a Sharpe ratio almost identical to that of the benchmark. Again, however, the excess return and Jensen's alpha results are not statistically significant. In terms of risk measures, three of the four funds displayed lower standard deviations relative to the market, and all four funds displayed betas below one.

Based on these results, in general, we cannot reject the hypothesis that there is no statistically significant difference between SRI mutual fund investing performance and conventional (non-SRI) investing performance in Canada.

This empirical study sought to introduce some rigour into the evaluation of the performance of Canadian SRI mutual funds by examining their returns with some well-recognized performance metrics and statistical tests, as opposed to simply examining gross returns.

The results suggest that those who engage in SRI through investing in Canadian SRI mutual funds, on average, are neither giving up anything nor gaining anything in terms of financial returns. However, it appears that the screens may actually decrease risk exposure, although such a conclusion depends on the extent to which the funds were fully invested in equities. In fact, as of December 2000, two of the funds (Investors Summa and Ethical Growth) had over 15% of the fund in cash while another two (Desjardins Environment and Clean Environment Equity) were virtually fully invested in equities (less than 2% in cash).

Proponents are likely to argue that since there is no difference in financial performance, those who engage in SRI are in fact better off - they are giving up nothing in terms of financial returns, plus they can sleep better at night knowing that their investment strategy is making the world a better place.

As the popularity of SRI continues to grow, as more SRI investment alternatives become available, and as the JSI and SRI mutual funds in Canada establish more of a history from which to draw data, the means of comparing the performance of SRI to conventional investing will be enhanced. At this relatively early stage, it appears that investing for the soul may not hurt the bottom line, particularly when risk exposure is taken into account.

1 For more information on socially responsible investing, see www.socialinvestment.ca (website of the Social Investment Organization) and www.socialinvest.org (website of the Social Investment Forum).

2 Social Investment Forum. 2000. "Fund Performance Update: Socially Responsible Funds Continue to Get Top Marks in 2000." News Release, Social Investment Forum, July 26, 2000.

3 Israelsen, Craig and Deanna L. Sharpe. 1999. "Social Butterflies: After Adjusting for Loads and Taxes, Socially Conscious Funds Have Provided Comparable Returns Over the Past Five Years." Financial Planning, July 1, 1999.

4 Hamilton, Sally, Hoje Jo, and Meir Statman. 1993. "Doing Well While Doing Good? The Investment Performance of Socially Responsible Mutual Funds." Financial Analysts Journal, vol. 49, no. 6 (November/December): 62-66.

5 Statman, Meir. 2000. "Socially Responsible Mutual Funds." Financial Analysts Journal, May/June: 30-39.

6 According to globefund.com, the relevant index of comparison for each fund is the TSE300. There is always the possibility that some of the funds at times may have sector characteristics that differ from the overall market index. For example, some SRI funds may tend to invest in high-tech firms that might have done particularly well over the period of study. As a check, we examined the major holdings of each of the funds as of the end of December 2000, as reported by globefund.com. In each case, the largest holding was a major Canadian bank stock.

7 There is always the possibility of a "survivorship bias" in a study like this. However, we are not aware of any SRI funds that were available at the start of the sample period but are no longer available.