Examining Risk vs. Reward
In our last post, "What is Asset Allocation?", we described 3 different asset classes (equities, fixed income, and cash) as well as the importance of the risk vs. reward relationship in the security selection process, specifically how different asset class mixes form different risk/reward profiles. Let's examine that relationship a little further by answering the following questions:
- how do we define risk?
- what are the historical averages of risk for different asset classes?
- why does looking at risk matter?
Disclaimer: investing involves many kinds of risk. This post will not be an exhaustive attempt to explain all risks associated with investing, but can serve as a general overview or framework for a discussion about risk.
Defining Risk in Equities (Stocks)
When discussing risk with equities, most often people think of headlines like "Dow Down 700 points!" or "Trade Wars Weighing on Stocks!" However, these are only examples of many variables that can impact returns. Most investors have no trouble readily identifying (what I'll call) the causes of risk, such as changes in GDP, unemployment trends, news headlines, or a surprising earnings report. Despite their newsworthiness, these causes are often hard to quantify and usually don't apply equally to different equity securities. For those reasons, most asset managers attempt to define risk not by the variables themselves but by measuring the effects that those variables produce. As each equity security reacts to news differently, the overall affect on the price can be measured and standardized. The most widely-used measure to define risk (variability) in equity securities is called Standard Deviation.
To explain standard deviation, below is an example of two ways I can drive to work each morning, Route A (the freeway) or Route B (side streets).
Route A - Freeway
Avg. commute: 8 minutes
Variability: 5 - 20 minutes
Causes of variability: Traffic, accident, debris, road rage
Route B - Side Streets
Avg. commute: 10 minutes
Variability: 8 - 12 minutes
Causes of variability: Traffic, bus stops, trash days, timing the lights
The average time to drive to work on Route A is 8 minutes, but the range of possible outcomes is between 5 minutes and 20 minutes. The variability is caused by a number of factors (or causes), but remember the causes don't matter much to us. It is the outcomes that we are focused on. The average time to work on Route B is a little longer at 10 minutes, but the range of variability is much tighter.
Calculating standard deviation: In simple terms, standard deviation is the measure of how much a sample of outcomes varies from the average. The tighter the outcomes are clumped together, the lower the standard deviation.The farther they are apart, the higher the standard deviation.
Applying standard deviation: If my goal is to consistently get to work on time, at first glance, the obvious choice could be the route with the fastest average (Route A). But when considering standard deviation (or the predictability of an outcome), Route B might make the less risky choice because knowing what time I will arrive (plus or minus 2 min) is better than the off-chance of being 12 minutes late.
The full story with standard deviation: One of the downsides of standard deviation is that without an adequate history of returns (at least 3 years), it's difficult to rely on the calculation. The plus side is that standard deviation is a simple way to standardize risk across the spectrum of equity investments, so you can compare one equity investment's standard deviation to another with relative ease. It's also important to note that with every new outcome, the standard deviation has the ability to change over time. It's possible there will come a time when Route A becomes the more reliable option.
The Big Idea: When it comes to standard deviation, stocks with a high variability in returns generally equals more risk, and stocks with low variability equals less risk.
Defining Risk in Fixed Income (Bonds)
Like equities, there are many variables that influence the returns of fixed income securities, and standard deviation can also be applied. Most often, variability in fixed income securities is from changes in the credit status of the issuer (lender), changes in inflation expectations, or (most often) the movement of interest rates. Credit quality is a big topic relating to bond risk that we'll save for another time. Since there has been a lot of discussion about interest rates lately, today we'll focus on the standardized way of assessing interest rate risk in fixed income securities, called Duration.
But before we get to duration, we must first understand the relationship of interest rates and bond prices:
Interest rates and bond prices have an inverse relationship: when one rises the other falls, and vice versa. Assuming the going rate of interest drops 0.25%, then when an ABC Company Bond yielding 5% (priced at $100) responds by falling to 4.75%, the price of the bond will increase to about $105. This makes sense, that a bond paying 5% would be worth more in a 4.75% world, so the price will increase until the excess yield advantage is gone.
Calculating duration: The process for arriving at a duration calculation is beyond the scope of this post. The result, however, is simply a number measured in years (i.e. 8.5 years) that is directly related to sensitivity to interest rate moves. Generally speaking, the higher the duration the higher the interest rate sensitivity, and the lower the duration the lower the sensitivity. Looking back to the ABC Company Bond see-saw, a long duration is like being further out on the end. The closer to the fulcrum (shorter duration) the less sensitivity, while the longer ends experience more change for the same move in rates.
Applying duration: Duration can be really useful if you wanted to know HOW MUCH a fixed income portfolio might gain/lose when interest rates move. For instance, Fixed Income Portfolio A has a duration of 6.0 years. A 1% increase in rates will typically decrease the price of the Portfolio A by 6%. And if interest rates fall 1%, the price of Portfolio A will typically increase by 6%. (Keep in mind: the relationship of interest rates to bond prices is inverse)
Another way duration is helpful: Fixed Income Portfolio B with a duration of 2.3 years will generally be less impacted by interest rate moves than Fixed Income Portfolio C with a duration of 8.0 years. Which takes us to the next point...
The full story with duration: Bond portfolios with a long (high) duration have more interest rate risk. Also important to note: duration is not the same as maturity (when the bond comes due or gets paid off). They are similar and often get confused.
Historical Averages of Risk for Different Asset Classes
Now that we have an understanding of risk, it's time to add some context. In the chart below (produced by Morningstar® and Ibbotson®SBBI®) you will find the historical (1926-2018) average for Returns (Geometric Mean) and Risk (Standard Deviation) for our 3 asset classes
For equities, we notice that Small-Company stocks have the highest standard deviation (31.6), with Large-Company stocks at (19.8). Now compare that to fixed income with Long Term Government bonds (9.8), and Long Term Corporate bonds (8.4). Then Treasury bills (which is similar to cash at 3.1).
Past performance is no guarantee of future results. This is for illustrative purposes only and not indicative of any investment. An investment cannot be made directly in an index. © Morningstar. All Rights Reserved.
This confirms our traditional views about stocks generally being more risky than bonds. But this hasn't always been the case. In the chart below you will find the averages for 5-year rolling periods for each asset class. Notice how there were periods where Large stocks were riskier than Small stocks, LT government bonds were riskier than Large stocks, and even Treasury bills (cash) were riskier than LT government bonds.
Past performance is no guarantee of future results. This is for illustrative purposes only and not indicative of any investment. An investment cannot be made directly in an index. © Morningstar. All Rights Reserved.
On the whole, it looks as if stocks are reaching a relative low in volatility (coming down from left to right). But notice that at the same time bonds are coming off the peak of historical volatility. Over the last 5 years, the bond market has been experiencing just as much (or even more) volatility than stocks!
So what? Why does risk matter?
The risk vs. reward relationship is not linear. The riskiest asset class (historically) has been Small stocks, but a standard deviation of 50% higher than Large stocks has not always resulted in 50% more reward. The assumption of "more risk, more reward" is misleading. The better question is "how much risk for more reward?"
That relationship can change over time. There are times where taking (perceived) less risk in LT bonds has actually resulted in more variability than stocks. As frustrating as it sounds, risk is a moving target, and as we mentioned in our last post, simply setting and forgetting your asset allocation might deviate outside your initial risk parameters. It's a good idea to check in on the standard deviation of equities and duration of fixed income investments every once in a while.
Looking at historical volatility helps keep your portfolio in perspective. There may come a time when stocks (again) reach historically low levels of volatility. Does that mean we should expect it to continue over the next 20 years? Maybe. But most likely it will change.
Risk can be counter-intuitive. Investors might think they can escape volatility by going to cash, which historically has been able to keep pace with inflation. Looking back to the SBBI Summary Statistics Chart, the long term inflation rate has been 2.9%. Today, ST Treasury Bills are hovering just above 2% with the expectation of going lower. If that continues for a longer period, then keeping too much cash in your portfolio could be trading one risk (volatility) for another (losing value by not keeping pace with inflation).
Bringing it all together, risk is one of the most important conversations we have with our clients. With so much focus on the reward, most clients don't know how much risk they are taking on to achieve it. Getting below the surface to understand the risks can help set the appropriate return expectations and can keep you invested when things take a turn.
If you would like us to help you understand the risk in your portfolio a little better, or have any comments you'd like to send our way, please don't hesitate to email us at info@curtisadvisory.com.