In the never-ending debate over the merits of active and passive investing, proponents of the latter argue that active management is a costly and challenging endeavour, because it is extremely difficult to find active managers that consistently deliver outperformance on an after-fees basis. The passive approach appeals to investors whose objective is to mimic or replicate market returns by using low-cost investments.
Supporters in the active camp, on the other hand, believe that in order to achieve above-market performance, one must do things that are different from the market. A truly actively managed fund can give investors the chance to beat the market and to offer protection in down-market environments.
There are merits to both, and it doesn't have to be all or nothing for either approach. Somewhere in the middle lies the Active-Passive approach, an alternative that combines the strengths and best qualities of both approaches into one strategy. The essence of the active-passive approach consists of two main elements: 1) investors need to identify markets where skillful managers have the ability to outperform the market, and 2) they must evaluate their own ability in picking outperforming active funds.
In developing a decision framework to blend active and passive investing into a single investment strategy, Morningstar Investments Management (MIM), the investment consulting division of Morningstar, conducted a study evaluating managers' ability to produce positive excess returns after accounting for fees and adjusted for risk for 13 frequently used asset classes in Canada over the past 23 years.
In order to identify which categories offer actively managed funds the greatest potential to deliver excess returns over a given period, we carefully analyzed the historical returns patterns and characteristics of the funds in a given category. We chose alpha as a measure of manager skill on a risk-adjusted basis, instead of the more commonly used annualized excess returns. This is because simple excess returns fail to distinguish the different risk profiles among funds within a category. Another shortcoming of the excess return approach is that it assumes a common performance standard (usually a broad market index) is a "good fit" for all funds in terms of capturing the range of securities in which each fund can invest, although this might not be an appropriate assumption even for funds that are in the same category.
Alpha can be calculated against any benchmark. In order to facilitate meaningful comparisons, it should be calculated against a benchmark that is an accurate representation of a fund's true style. In this study, we created a best-fit custom benchmark for each fund by regressing the fund's historical net-of-fees returns against a list of indices. Once we have determined the custom benchmark for each fund, we can start calculating alphas.
Instead of just using one alpha for the entire performance history of a fund, we estimated monthly alphas going back to January 1990. This essentially gives us a time-series of alphas that forms the basis of the distribution. Using the distribution of alphas has a number of benefits. First, it provides the entire range of historical alphas instead of just point estimates (i.e. averages). The distribution also reflects the expectations for managers' skills in a category. For instance, if the majority of the observations in a particular category lie above 0%, this indicates that actively managed funds not only perform well as a group, but that investors should also have higher confidence in their ability to pick top-tier funds. On the other hand, if the bulk of the observations fall below 0%, then active funds are not adding any value and investors would need to do more research and due diligence work to uncover outperforming funds.
Our research has shown that managers have demonstrated more success in generating alpha in some categories than others. The table below shows the distribution of alphas for 96 funds in the Canadian Equity category, which are mostly large-cap funds, and 60 funds in the Canadian Small/Mid Cap Equity category, using monthly return data from January 1990 to December 2013.
Over all monthly observations during that period, positive alphas only occurred 29% of the time in the Canadian Equity category compared to 60% in the Canadian Small/Mid Cap Equity category. This shows that, as a group, Canadian small-cap equity funds performed better with active management than their large-cap counterpart.
| Distribution of Alpha: Jan. 1, 1990 to Dec. 31, 2013 | |||||
| Alpha Range | Canadian Equity | Canadian Small/Mid Cap Equity | |||
 |
|||||
|
|||||
| < 0% | 15% | 17% | |||
| = 0% | 56% | 23% | |||
| > 0% | 29% | 60% | |||
|
|||||
| Source: Morningstar Investments Management | |||||
|
|||||
Graphically, the dispersion of alpha shows most observations for Canadian large-cap equities are centered around the 0% alpha range, whereas for Canadian small-cap funds alphas are more evenly distributed and slightly skewed toward the positive side. What this means is that for an investor who has absolutely no skill in picking winning fund managers, he or she would have a better chance of selecting a fund that generates positive alpha on a random basis in the Canadian small-cap equity category than in the Canadian large-cap equity category.
 |
|
 |
|
|
|
|
|
|
The analysis presented here is designed to provide an alternative approach in implementing long-term strategic asset allocation strategy, helping investors identify markets where they are being rewarded the most to be active. The construction of an active-passive investment strategy that is backed by defensible research is a complex process that requires a lot of time and resources. Investors who are interested in constructing their own active-passive portfolios should consult a qualified financial advisor.
