A client brought up the concept of being too diversified recently, and it gave me pause. My initial reaction was that there is no such thing . Diversification is a good thing. It allows you to minimize risks inherent in individual investments (company gets sued, senior executive leaves, “next big thing” flops, etc.) and take the risks you should be focusing on. As I’ve said before, diversification is the only free lunch in finance. But the client’s question wasn’t really about diversification – it was about whether we are taking the right risks.
What Risks Don’t We Want to Take?
This has got me thinking about some of the risks that we want to take, and the risks that we do not want to take. We’ve covered the risks that make sense in the context of our portfolios and we want to take, but we haven’t covered the types of risk that don’t make sense.
We ask three questions when looking at a risk premium:
- Is this actually a risk factor?
- Does this work in the context of our portfolios?
- Can the risk premium be captured efficiently and effectively?
We use these (mainly the first one) to focus in on risks we think actually belong in client portfolios. We want to take risks that increase the expected return of the portfolio as a whole, and leave out everything that doesn’t. Let’s break these down.
Is This Actually a Risk Factor?
Basically all of the hot new investments you hear about should probably be avoided. To be blunt, you shouldn’t be interested in most of what the investment industry is shilling. They may be able to point to really great past returns, but why should we think those returns will persist? Any number of strategies worked for years and just stopped working one day. That’s known as the end of a lucky streak. If we don’t understand why something is getting the returns it does or it seems like random noise, we stay away.
We frequently get questions about investing in commodities, and they fail this test. Commodity returns are basically random noise around a relatively low average return. Since 1992, the Bloomberg Commodity Total Return Index has had an average annual return of 5.29%, and an annual standard deviation of 18.67%. To put that in perspective, over the same time period the CRSP 1-10 Index (the best measure of the total US stock market) had an annual average return of 11.34% and an annual standard deviation of 18.54%. So commodities carry about the same risk as investing in the stock market and return significantly less than half of the average return.
If you step back and think about what you are investing when you invest in commodities, this makes sense. Commodities are inputs, not businesses that can grow and thrive. Commodities move around based on two things: how valuable they are in making something else, and how much investors think they’ll be worth in the future. You’ll notice there is nothing about profits or creating value or distributing cash flows to shareholders.
The industrial demand is (moderately) stable, so let’s think about the part based on investor sentiment. This is a perfect example of what’s known as the greater fool theory. Investors think it is valuable because someone else will buy it for more money in the future. As soon as that belief is questioned, the whole thing comes crashing down. In essence, it’s a continual bubble-and-bust cycle with commodities. We don’t want to play that game, and more importantly, we don’t think it helps our clients reach their financial objectives.
Does This Risk Work In Our Portfolios?
If a risk makes it through the first screen, we’re still not quite ready to invest. We need to make sure it improves the risk characteristics of the total portfolio. It should either give us a slightly higher return with the same risk or the same return with a slightly lower risk. This can be done in a number of ways, but the two big ones are to simply have a better individual risk profile or to have a low correlation with the rest of the portfolio.
The individual return aspect is pretty straightforward. If something has the same amount of risk with a significantly higher return, we’re going to be pretty interested. Something with a low correlation with the rest of the portfolio moves differently than the rest of the portfolio, meaning it can actually reduce the total risk in the portfolio – even if it’s more risky than the total portfolio by itself. That also peaks our interest.
High yield bonds would be screened out at this level. “High yield” is a fancy phrase meaning “junk.” We understand why they have higher returns than other bonds – they’re bonds from companies that are less financially secure and therefore more likely to fail to pay back the bonds. That makes sense. But the way this plays out in the market just doesn’t make it a good fit for the portfolio. Because the value of these bonds moves reasonably closely with the financial fortunes of the companies that borrowed the money, they actually act a lot like stocks.
From July 1983 to July 2015, the Barclays US Corporate High Yield index had a correlation of 0.6 to the CRSP 1-10 Index. To put this in perspective, the Merrill Lynch 1 Year US Treasury Note Index had a correlation of -0.018 with the CRSP 1-10 Index. This means that adding high yield bonds to our portfolios doesn’t really do that much in terms of reducing the amount of risk in the portfolio. There is nothing inherently wrong with high yield bonds, they just doesn’t fit how we build our portfolios, so we avoid them.
Can the risk premium be captured efficiently and effectively?
If a risk makes it through the other two screens, then we start looking at how we can capture it. It’s one thing to be able to point to a whole bunch of research showing a risk factor has everything we could ask for, but it’s another to go into the markets and actually capture the risk factor. A number of reasons make it difficult to capture a risk premium – it could be expensive to trade the securities related to the risk factor, it could involve so much turnover that taxes and commissions start eating up the premium, or the premium could just be incredibly volatile.
This is as much looking at the risk itself as the market and funds trying to capture the premium. It’s also important to point out that something really hard to trade in the past can be very easy to trade now.
We also evaluate momentum as a risk factor. There is undoubtedly a risk premium there, and it seems like it could be a really good fit for our portfolios, but capturing the risk premium involves a lot of portfolio turnover, incurring numerous of transaction costs and taxes. Both of these eat into the portfolio returns and make the premium less attractive. A lot of people are looking into how to trade this premium effectively, but we don’t think they are quite there yet – though it is probably only a matter of time.
We’re constantly looking for ways to improve our portfolios. This includes looking for funds that capture the risks we want more efficiently, but also looking at the risks we actually want to build into our portfolios.