Mutual fund risk ratings have long been an area of contention for fund providers and confusion for financial advisors and their clients. Fund companies must disclose the “risk” of their fund in the prospectuses and fund facts documents that they are required to provide. But nailing down a universally acceptable methodology for assigning “risk” has proven elusive. In an effort to standardize the way in which fund companies assign risk ratings to their funds, the Canadian Securities Administrators (CSA) in December 2013 issued a request for comment on its proposed mutual fund risk classification methodology
. The CSA’s idea is a simple one and should be easy enough to execute. But in the mutual fund industry, nothing is really ever “simple” or “easy.”
Fund managers, regulators, investors, advisors, not to mention battalions of product sales and marketing folks, each have their unique perspective on what a standardized methodology should look like. Complicating matters is the fact that some stakeholders are more concerned with the impact that the new risk ratings would have on their competitive and financial position rather than the effectiveness of the methodology used to develop the ratings. Before we get into the contentious issues involved and the debates the process has spawned, let’s a take a look at the proposal and see how it differs from current industry practice.
The CSA is proposing that fund companies use standard deviation (SD) to determine the risk ratings for mutual funds. Standard deviation is a measure of volatility or price risk. The higher the SD value for a given fund, the greater its historical monthly increases or decreases in price. The CSA has proposed using SD because it is arguably the most well-known and easily understood risk metric and is currently already used by a large number of fund companies in assigning risk ratings.
The CSA is proposing that companies use 10 years of history to calculate the SD. In cases where a fund has less than 10 years of history, it must use a suitable proxy to backfill history. The risk rating is then set according to where the SD lands on the proposed scale (see below).
Until now, fund managers have been free to use any method they wish to determine the appropriate risk rating for their funds, with the caveat that they must disclose their methodology in the prospectus. Methodologies differ throughout the industry, but it is believed that the vast majority of managers currently use the “IFIC Methodology.”
The Investment Fund Institute of Canada (IFIC) has set forth risk-determination guidelines in its publication Voluntary Guidelines for Fund Managers Regarding Fund Volatility Risk Classification
. It recommends that fund companies compare the SD of the fund (or proxy benchmark) with a prescribed index associated with the fund’s Canadian Investment Funds Standards Committee (CIFSC)
category. If it does not “differ materially,” then the risk rating should be set based on the risk level of that CIFSC category. If it does differ materially, then the risk rating should be based on the average of the rolling 3-year and/or 5-year SD.
The table below illustrates the differences in the risk categories and thresholds between the IFIC recommended guidelines and the proposed CSA guidelines. As you can see, there are some significant differences in the bands for each risk category, as well as the addition of a sixth category under the CSA proposal.
The CSA allowed for a comment period that closed in March 2014, and there was plenty of feedback. While most respondents agreed that the industry needed a standardized methodology, there was a wide range of opinion from the various industry participants on what the standard should look like.
Here’s a look at some of the main sticking points with the proposed methodology as well as some of the significant differences between the CSA proposal and the IFIC Guidelines.
Time period – How much history to consider?
The length of the time period used to calculate the SD is a tradeoff between consistency and relevance of data. The IFIC Guidelines use 3-year and/or 5-year SD, while the CSA proposal uses 10 year SD. Only around 20% of mutual funds have been around for 10 years, while only about 4% of exchange-traded funds (ETFs) have a 10-year life. This means that under the CSA proposal, the majority of funds will have their risk rating based on a proxy. This is not a big problem for funds that track indexes with a long history, but it becomes more of an issue for actively managed funds or funds that track new indexes that do not have a 10-year track record. For comparison, 40% of mutual funds have at least five years of history, while 55% have at least three years of history.
One of the benefits of using a 10-year SD is that it eliminates much of the variation in the measure itself. This means that risk ratings should be more consistent, even if the market goes through extended stretches of either high or low volatility, and eliminates the need to adjust the SD bands periodically. Furthermore, mandating a 10-year measure removes the potential biases that can be caused when companies are free to choose the time period on which to base their calculations.
In contrast, using 3-year and/or 5 year SD under the IFIC Guidelines allows for the risk measure to capture recent volatility trends in the market and might follow more closely with what investors actually experience.
Risk scale – How many risk categories and what are the thresholds?
The CSA recommendation proposes a six-category risk scale; the IFIC scale is five categories. The CSA added a “Very High” category to capture extremely volatile funds. Additionally, the thresholds being proposed are very different than the IFIC scale that many companies currently rely on. As a result, using the proposed thresholds would instantly shift a large number of funds up at least one risk band.
The initial risk rating changes that result will certainly be disruptive for the industry and confusing for investors, particularly since nothing has changed about a given fund or the way it is managed. To further complicate the situation, many risk-management systems used by dealers have incorporated the current five-category scale into their Know Your Client (KYC) rules.
Moving funds up a risk band or two will mean that some funds will no longer be “suitable” investments for clients based on their risk tolerance, and many portfolios will have to be adjusted accordingly. On the other hand, the argument can be made that basing suitability on investment-specific risk ratings rather than taking a holistic portfolio view is a flawed approach to begin with.
Monitoring – How frequently should the ratings be updated?
The CSA has proposed monthly monitoring of the risk ratings with prescribed rules on when a ratings change must be made. Under the IFIC Guidelines, a review must be undertaken “at least annually,” and as a result risk rating changes usually happen when companies are renewing their Fund Facts documents. Mid-year revisions are rare and typically only occur if there has been is a material change to a fund where the associated risk has either increased or decreased significantly.
Opponents argue that monthly monitoring is excessive and that yearly monitoring would be more appropriate. The counter-argument is that this data should be readily available to fund managers, who should be monitoring the numbers anyway. Under the CSA proposal, if the monthly calculation shows that the risk band has changed by two levels up or down, the risk rating must be changed on the Fund Facts document within 10 days of the last monthly calculation. Otherwise managers will have to calculate the average risk category over the past 12 months to determine whether or not a change is needed.
Frequent changes to risk ratings are certainly not desirable, and risk ratings should be as consistent as possible. But at the same time, investors should be alerted as soon as possible to shifts in volatility rather than having to wait until an arbitrary date that the company uses as its fiscal year end.
While these are the main issues at hand, many other details are still being debated. And the industry is anxiously awaiting the CSA’s next proposal to see what changes, if any, have been incorporated based on the comments received. Whatever methodology ends up being mandated, you can be sure that almost no one will be completely satisfied. But to the extent that Canadians investments are being driven by these risk ratings, standardization and consistency across investment products is a much-needed and welcome development for everyone involved.
Brian Bridger, CFA, FRM, is Director, Analytics & Data at Fundata Canada Inc.
Reid Baker is Manager, Analytics & Data at Fundata Canada Inc.
© 2015 by Fundata Canada Inc. All rights reserved.