Problems arise with fund risk ratings

By Brian Bridger

July 4, 2019

It has been over two years since managers first began calculating investment fund risk using the CSA Mutual Fund Risk Classification Methodology. (See my previous article for a review of the methodology and some of the initial concerns.) Two important aspects of the methodology are now leading to inconsistent and potentially misleading risk ratings. The first is the static standard deviation (SD) bands used to assign risk ratings. The second is “upward discretion” afforded to managers. Let’s take a look at both of these issues.

Standard Deviation Bands

Under the methodology managers must calculate a 10-year SD using proxy data to backfill if necessary, and then assign the risk rating based on the following table:

When developing this methodology, the regulators argued that 10-year volatility levels were consistent over time, thus eliminating the need to shift the bands based on prevailing market conditions. However, the low volatility environment of the past three years has cast serious doubt on this assumption, something that was predicted over two years ago.

Since the CSA established the SD bands in 2016, the average SD for each category set out by the Canadian Investment Funds Standards Committee (CIFSC) has gone down. In fact, for 18 of the 34 categories that had a 10-year track record in May 2016, the average fund has shifted down one SD band. For example, the average Natural Resource Equity fund has gone from an SD of 22% (High risk) down to 17.5% (Medium-to-High risk) in the past three years. The table below illustrates how other categories have been affected.

For more than half of funds in Canada, the static SD bands now point to a rating that does not accurately reflect the riskiness of the investment.

Upward Discretion

The intention was good: Give managers upward discretion and they would voluntarily increase risk ratings (or rather not lower risk ratings) if a period of low volatility led to inappropriately low ratings. The problem is that this theory only works in practice if all managers are acting in unison. Otherwise you end up with similar funds having different risk ratings.

In 2017 there were around 210 risk rating decreases, compared with only 70 increases. In 2018, the numbers were closer to 200 and 80, respectively. Over these two years, risk-rating decreases represented over 70% of all rating changes, and around 7% of funds had a risk-rating decrease each year.

So far in 2019, over 145 funds have had a risk rating decrease and only five have had increases. The number of decreases now represents over 12% of funds, and the ratio of decreases to increases is 29 to 1. At this pace, we could see another 300 risk-rating decreases by the end of the year. Clearly, more and more managers are taking advantage of the low-volatility environment to reduce their fund’s risk rating rather than using upward discretion.

The following data are based on funds with a 10-year track record, which includes 1,531 funds (approximately 40% of the 3,856-fund universe). We’ll start by looking at the Global Equity category.

Global Equity is one of the largest CIFSC categories and contains funds employing various styles and investment approaches, each with varying degrees of risk. So it is not a surprise to see a wide range of volatility numbers. Nevertheless, we can make some interesting observations.

The first is the lone High-risk fund. It has a 10-year SD of 10.4%, just slightly above the category average of 10.3%, which translates to a risk rating of Low-to-Medium. This manager is clearly using upward discretion, moving the risk rating up three levels! But what about the rest of the managers?

Five funds are rated Low-to-Medium, despite having a higher SD than the average Medium-risk fund. Similarly, four Medium-risk funds have a higher SD than the average Medium to High fund. This illustrates the problem with giving managers the discretion to increase risk ratings. Some will use it and others will not, leading to similar funds having different risk ratings.

U.S. Equity is another category with a wide dispersion of risk ratings. The table below illustrates the breakdown.

Some managers are using upward discretion, as is the case with the two High-rated funds whose volatility equates to Medium risk. But clearly, many managers are not doing this and have simply lowered their ratings based on the SD bands. For example, nine funds rated Medium have a 10-year SD above the average for High-rated funds, which are two risk bands higher.

There are 698 funds out of 1,531 that have a higher risk rating than their 10-year SD would suggest. But this is not necessarily an accurate reflection of the number of managers using upward discretion. The reason is that risk ratings are normally updated only once per year. Given this, it is possible that a fund’s 10-year SD has recently dropped an SD band, and the manager will lower the rating on the next Fund Facts or ETF Facts. By looking at changes in the 10-year volatility numbers since a fund’s last risk-rating date, there could be close to 400 funds that fall into this category. If we extrapolate this for the entire universe, we are talking about close to 1,000 funds that would be eligible for risk-rating decrease.

Static SD bands are leading to lower risk ratings across all investment fund asset classes. Meanwhile, upward discretion is causing inconsistent ratings between products. The last thing the industry wants is another round of consultations on risk ratings, but you have to wonder how long the regulators will let this continue.

Standardizing the risk-rating methodology was a long journey, but in the end it was a huge win for investors and other industry participants looking for reliable and consistent product data. Now it’s up to the regulators to make sure the risk ratings stay relevant. Otherwise, what was the point of all this?