By Dann Ryan, CFP®
Senior Manager, RCL Advisors, LLC
The debate of active vs. passive management is unlikely to see a conclusion in the near future. However, if you adopt a neutral viewpoint you can quickly see applications for both strategies. In investment management practice, by being objective about your own search criteria, it may be possible to create a quantitative search process that enhances your success for qualitative active manager selection. These selection criteria need not be particularly restrictive, but rather just reflective of your actual investable universe. For many typical individual investors this universe could be expressed by the mutual funds which are covered by Morningstar’s database.Using this as my source and for purposes of my analysis, I’ve employed the below search process:
Fund Share Class Criterion
To limit the duplication of manager data, one share class per fund was selected based on the following preference:
- Institutional Class
- Lowest Annual Report Net Expense Ratio
- Lowest Maximum Management Fee
- Oldest Performance Start Date
Obviously expense ratio impacts net-of-fee to manager performance, so it is important to be selective with the share classes which are included in the investment universe. If you are an institution or large advisor with access to preferential share classes, this should be reflected. Conversely, if you are an individual you may be limited to investing in the more expensive retail share classes.
Fund Universe Selection Criteria
The funds were then searched and screened by the following criteria:
- Morningstar Category (US Open End Mutual Funds)
- Minimum of 5 years performance history
- Minimum of 5 year R-squared of 90 to the relative benchmark
- No Index Funds
The Decision to Use Alpha
Alpha as a statistic attempts to quantify an investment manager’s ability to add value adjusting for the systematic risk (beta) of the asset class. Inherent in this performance is the fund’s expenses, which are critical to the client’s experienced benefit. By this definition, index funds would be expected to have a negative alpha. Furthermore, by restricting the search to funds with high R-squared, you are helping to increase the integrity of the alpha calculation by improving the accuracy of the beta. However, it is important to note that alpha is not synonymous with outperformance, but rather excess risk-adjusted returns. This definition seems sufficient, as there are certainly instances when you wish to implement less than market risk strategies for a given asset class.
After defining the investment universe by applying the share class selection criteria described above, I then pulled rolling twelve month time periods stepped monthly for the last 10 years ending December 31, 2013. This attempts to identify the likelihood of any one fund adding value over a 12 month period, given reasonable adjustment for persistence of returns.
As a secondary test against time period anomalies, I also ran these criteria for the five years prior ten year period ending December 31, 2008. Admittedly, this introduced a small amount of survivorship bias as it excludes managers that may have gone out of business since, but would have been part of your investable universe five years ago. Although this period was one of great outperformance for growth style managers, I otherwise found overwhelmingly similar conclusions to the data from the most recent ten years.
The process attempts to find asset classes in which managers had the highest incidence of generating alpha. As an investment professional with an established due diligence process, including qualitative factors, I believe that picking an active manager that could succeed at providing benefit to my clients in that space.
The results were as follows for the following categories:
|Foreign Small/Mid Blend||MSCI EAFE Small Cap NR||14||1,422||62.94%|
|Small Value||Russell 2000 Value TR||57||6,356||61.34%|
|Foreign Large Value||MSCI EAFE Value NR||67||6,975||61.16%|
|Small Blend||Russell 2000 TR||140||15,199||60.02%|
|Foreign Large Growth||MSCI EAFE Growth NR||47||5,199||59.55%|
|Small Growth||Russell 2000 Growth TR||151||17,034||58.31%|
|Foreign Large Blend||MSCI EAFE NR||140||14,512||53.18%|
|Diversified Emerging Mkts||MSCI EM NR||85||8,527||50.05%|
|Large Value||Russell 1000 Value TR||210||23,485||47.86%|
|Mid-Cap Growth||Russell Mid Cap Growth TR||145||16,585||46.05%|
|Large Growth||Russell 1000 Growth TR||314||35,535||44.87%|
|Large Blend||Russell 1000 TR||267||29,452||43.13%|
|Mid-Cap Value||Russell Mid Cap Value TR||69||7,399||42.92%|
|Mid-Cap Blend||Russell Mid Cap TR||64||6,905||36.02%|
Source: Morningstar Direct
Foreign equity managers offered much more alpha generation opportunities than US equity managers. This may be attributed to the United States equity markets are some of the oldest and most efficient and this reduces managers opportunity sets. This international benefit was most pronounced for managers in the Mid-Small capitalizations; however the same size of funds was also lowest here. Perhaps this is attributable to market inefficiencies or simply immaturity of the asset class.
In general, there seemed to be a noticeable advantage to mangers who employed a growth/value bias as opposed to those with a blended approach. With the exception of Small Cap Growth, style-specific plays always produced more benefits than their blended counterparts.
The US Mid-Capitalization manager space provided very little opportunity for value addition. Especially for blended strategies, a strong case can be made that you are best to index this space.
Emerging Markets have typically been a space in which a strong argument has been made for delineation amongst investments. Certainly there is a great breadth of variance in the economies of the respective countries, and conceivably this offers managers a great opportunity to add value playing on macro themes and valuation discrepancies. However, by this alpha generation metric there simply doesn’t seem to be that strong of a case for active management.
This article tends to describe a process to give you confidence in the ability to identify value adding managers in categories where the majority (>50%) of time periods managers are generating alpha. Also, it identifies spaces where alpha generation is much harder to produce and therefore you might just obtain the asset allocation diversification benefits of these classes through a low-cost index mutual fund of ETF.
Certainly as clients investment constraints change, so does their investable universe, and so the selection benefits may change drastically. Therefore my results may not be appropriate for everyone’s practice. It’s important to be accurate in your search criteria so not to overstate your success, or else you may likely be better off to employ passive management entirely.