This post is part of an ongoing series on the Attain Capital blog that seeks to help investors understand the various metrics we use to evaluate managers. Stay tuned for future pieces!
* * * * * * * *
We recently did a post that explained everything you’d want to know about the Sharpe Ratio: how it’s calculated, why it matters, and how you should use it. The short version is that the Sharpe Ratio is a stat which measures returns relative to volatility. In looking back at that piece, however, we failed to mention one of the big failings of the Sharpe ratio (we mentioned three others).
That failing is that the Sharpe ratio considers all volatility in an investments’ returns a bad thing. But is a large upside return really a bad thing? Some may say yes, as it means they could have an equally as large negative return. But for investments with a long volatility profile where they can earn outlier returns by letting profits run, punishing programs for those outlier returns (because they increase the investment’s overall volatility) doesn’t seem to make a lot of sense.
So how do you solve for this deficiency? Enter the Sortino Ratio, which is essentially the same thing as the Sharpe ratio – measuring returns over volatility- but which only considers downside volatility.
What is it? The Sortino Ratio is a risk adjusted return statistic which measures an investment’s return per unit of risk, with risk equal to the standard deviation of negative returns.
Sortino = (Compound ROR – risk free ROR) / (Standard Deviation of Negative Returns)
The Sortino Ratio, developed by Director of the Pension Research Institute Dr. Frank A. Sortino in 1980, can be viewed a modification of the Sharpe Ratio that treats risk only as the downside volatility in an effort to solve for the Sharpe’s problem of penalizing programs for positive outliers, as the Sharpe Ratio penalizes both upside and downside volatility equally. This solves the bulk of our issue with the Sharpe Ratio’s limitations, but is not without its own problems.
Remaining are two of the problems we mentioned in discussing the Sharpe ratio – 1. That there is a lot more to risk than just volatility (even downside volatility), and 2. Using volatility assumes returns are distributed normally – which they aren’t. From the Sharpe post:
The largest issue of using volatility of returns, and more technically the standard deviation of returns, is that using such a calculation assumes a normal distribution of returns. That means the Sharpe ratio assumes returns are spread nice and evenly on a bell curve, like the heights of students in a grade school. Thing is, the financial world is anything but a nice smooth bell curve, with events which are deemed to happen once every 10,000 years on the bell curve happen once a year or more. Financial returns are not normally distributed, so using such a metric to gauge their performance is fraught with problems.
On final point – while the risk metrics you’ll see on the Attain website and others offer valuable insight to the investor, looking at a statistic in a vacuum is fairly worthless. For one, the Sortino Ratio, much like other ratios used, is only meaningful when compared to the Sortino ratio of another program. The higher the number, the better, but on its own, a 1.42 Sortino Ratio means nothing. Moreover, even when used to compare programs, as we’ve explained, the Sortino Ratio does not provide the full picture, so it is best used in conjunction with a variety of other metrics.