If there is such a thing as the holy grail of investing, it is probably to construct a portfolio that exhibits low volatility in combination with a high level of yield. While that apparently goes against the ground rules of investing which say that risk must be commensurate with returns, research in recent decades has shown that over the longer run, stocks with lower volatility have actually outperformed their higher risk peers. That opens up a world of possibilities.
But simply picking high yielding low volatility stocks is not going to do the trick, as I will outline below. While such a portfolio may beat other equities, it will still leave you with significant levels of risk and exposed to large losses in case of a 2008-style stress event.
Instead, in this article I will tell you how you can construct a better diversified portfolio that achieves high levels of yield using not only low volatility equities, but also other asset types.
The portfolio below achieves our aim of a high yield at an even lower level of risk, while also faring far better than the broader market in even the worst of the market downturns that we have seen. Naturally, there are some risks and caveats to this “having your cake and eating it” strategy and I will highlight those as well.
More than 5% yield at less than 5% volatility?
When talking about low volatility and high yield we should be aware that they are both very subjective terms, take the S&P500 today for instance:
- In its regular form its volatility is 15% at a dividend yield of 1.9%. In 2008 it returned -38%
- Its low volatility high dividend variant (S&P Low Volatility High Dividend Index) has a volatility of 13% at a dividend yield of 4.4%. In 2008 it returned -24%
The second index clearly improves on the former, but in case of a 2008 style crash it would still take you more than 5 years to recoup your losses on dividend yield alone, that is on the assumption that companies would even keep paying.
Most of the minimum volatility ETFs that are currently out there don’t have sufficient history to tell us how they would have fared in 2008, but if we look at the recent bout of market volatility (Feb 2018), we see that the iShares Edge MSCI Minimum Volatility USA ETF returned about -9.2% in a two week period. That raises my eyebrows a little.
Making things worse, I am hard pressed to call 13% a low level of volatility even in normal markets. If you invest $100,000 and I tell you that in a normal scenario by year end you’d have a portfolio between $87,000 and $113,000, would a low-risk investor be happy with that? Probably not.
So how can we further improve on this? The key is to search for other asset classes (or high yielding subsets thereof) that complement low risk high dividend equities and offer similarly high yield, but diversify some of the volatility away.
I will outline such an ETF portfolio below that delivers more than 5% annual yield (income) at less than 5% volatility by combining low volatility high yielding stocks from different markets, with bonds, preferred REITs and senior loans.
Per estimations from the risk model (Bloomberg PORT), in a 2008 like stress event it would take this portfolio less than 3 years to recover on yield alone, compared to 5 years or more for most low-vol high-yield equity portfolios, or well over a decade for a regular equity portfolio.
Why do investors like yield?
Total return = yield + capital growth
Total return consists of both capital growth and income or yield. Yield can be gotten from sources like bond coupons, stock dividends, rental income or indirectly through ETFs that distribute the yield they reap, and many investors view it as a current income stream.
They are not too concerned with markets moving up or down somewhat, as long as the longer term trend is up and yields keep being paid. After all, cash in hand is better than a position in a security that may drop in value any second.
With bank deposit rates in developed markets are still very low, bond yields having become compressed yet at the same time equity markets looking scary being at all-time highs, investors are increasingly looking to risky assets to provide the income that they are not able to find it in traditional low-risk places anymore. That’s the purpose of yield.
If I had invested in the S&P500 since 2000, my total capital growth would have been +81%, while my total return would be +156%. Roughly half of that can therefore be attributed to yield, and that is what we are searching for here
Stock dividends and bond yields are relatively stable, providing a cushion for volatility. If you hold a 5% dividend paying equity, then 5 years down the line you would have earned sufficiently to break-even with the occasional -25% stock decline.
While when it comes to equities, dividend paying stocks may actually be less risky than their peers as they tend to be more established companies, with bonds it is the other way around. Income from bonds is inherently related to risk (a lower credit quality means higher borrowing costs for firms, which investors can benefit from if the firms manage to stay afloat until their debt matures). Sound risk management therefore needs to be applied.
But low volatility does not necessarily mean low risk
Volatility, which is typically used as a proxy for risk, tells you something about the most recent risk level of an asset. It tells you how wildly an investment can swing up or down in normal market conditions, with 84% certainty.
Pre 2008 there were equities that exhibited lower volatility than their peers, yet when the market crash came they still sold off by 25% or more. That is not exactly low risk, but still not bad if you realise that the whole market lost nearly 40% in value around that time.
This is why I personally like to view the risk level of an asset in two ways: their current volatility, and the amount you could stand to lose in a true stress event. Markets are currently in a low volatility environment. This can lull us into thinking that some equities are “defensive”, “low volatility” and therefore offer equity-like upside without the downside.
Indeed, compared to their peers their risk-return trade-off is superior (which may be driven by anything from human behavior and preferences to sector-specific dynamics), but when a systematic crash hits markets you can still expect everything to go down the drain at the same time.
So simply picking a low volatility equity ETF is not going to cut it. The portfolio I am constructing here should have overall low volatility, but it should also not suffer too much during a stress event.
Therefore diversification is still key
Low volatility investing has generated superior risk adjusted returns over the long term. It goes against standard market theory that more risk is rewarded with more return. Behavioural biases, rebalancing effects and investor constraints are common explanations for the low volatility premium. (AON, “Factor Investing: What’s the big deal with low volatility equities?”)
Only high yield and superior risk-return characteristics within each asset class is not going to be sufficient. We want to increase the risk-efficiency of the portfolio and obtain high expected annual yields yet at the same time achieve lower overall volatility than simply the weighted average of the components. In addition, we want to minimise the damage the next market crash can do.
The chart above shows (March 2018) risk levels vs current rolling annualised 3 year returns for various asset classes. Note that this shows total return rather than yield, so it does not necessarily paint a perfect picture of which assets are high yielding (the yield on gold is precisely 0.0%), but it goes a long way.
What we are essentially looking for here are asset classes where volatility and returns are at least close to each other if not better, preferring lower volatility. Based on the past 3 years, S&P500 Low Vol High Dividend, EM Bonds, Preferred REIT and Global High Yield look particularly appealing.
So low vol high dividend equities are a given and going to be a part of our portfolio. What asset types exhibit low correlation with these investments? Again Preferred REITs and EM Bonds stand out.
So what assets should we invest into?
We therefore need to think about adding some other assets into the mix, preferably those that can offset some of the shock in stress scenarios, while still delivering a decent and sustainable yield. I scanned the ETF universe for funds across the asset classes that had the lowest 90 day volatility and the highest yield levels, and found good ETF candidates in all of the below asset types.
This incorporates the EM bonds, high yield bonds, low vol high dividend equity and preferred REITs that we had already picked up above. In addition, senior bank loans stand out as an asset class that adds both diversification and yield.
- Minimum volatility equity. Scientific papers have actually proven that minimum volatility equity works to improve risk-return characteristics, which is also obvious from the popularity of such products. During 2008 the 100 lowest volatility stocks of the S&P500 returned -24% while the index as a whole lost -37%
- High dividend stocks. High dividend stocks tend to have lower volatility (and there thus appears to be quite a bit of overlap with minimum vol equities), be more established business and help build up a cushion to cope with the next stress event
- Preferred stocks of REITs. Preferred securities issued by Real Estate Investment Trusts offer high quarterly dividends and low beta equity exposure. A risk is that these securities tend to be callable (which may lose you money if you buy at a premium to par) and increased interest rate risk. However correlations with other asset classes are lower, volatility is in between bonds and stocks and yield is high
- Senior loans. floating rate securities are notes issued by banks or other financial firms that typically have volatility higher than investment grade bonds
- Unconstrained bonds. Typical global fixed income funds have high duration and low yield, and practically only exposure to developed markets. Unconstrained bond funds are able to reduce much of the interest rate risk and add on a healthy dose of yield by investing in emerging markets or high yield corporates. Should in theory be able to weather any market conditions and have lower correlation with other asset types
Portfolio construction: 5 high yielding ETFs
- The suggested portfolio consists of 5 high yielding ETFs within low volatility equity, senior loans, high dividend equity, preferred stock of REITs and unconstrained bonds. Or at a higher level about 40% equity, 20% preferred equity and 40% fixed income like instruments. The correlation between all constituents ranges from -0.7 to +0.4, the entire portfolio is therefore highly diversified
- This achieves a volatility of 4.8% versus a yield of 5.2%, which is seriously impressive. However during a 2008 style crash, the portfolio could still lose some 16%. But then at its yield, it wouldn’t take you more than 3 years to recover (or to start building up the cushion today)
- I like this portfolio and reckon its a good addition to almost any other broader allocation, therefore I’d recommend an allocation of 10%-20%
|Min vol equity||iShares Edge Min Vol Europe Currency Hedged ETF||HEUV||0.28%||20.0%
|Preferred REIT||Infracap REIT preferred ETF||PFFR||0.49%||20.0%
|High Dividend||Global X Superdividend||DIV||0.45%||20.0%
|Senior loans||Highland iBoxx Senior Loan ETF||SNLN||0.55%||20.0%
|Unconstrained bond||JPMorgan Global Bond Opportunities ETF||JPGB||0.55%||20.0%
Risk, diversification and allocation
- Risk level: low
- Diversification: medium
- For risk and total return since initiation see Portfolios
- Probability of this theme playing out in the next 3-10 years: 50%-100%
8. Portfolio characteristics (full look-through, from USD perspective)
- Dividend yield: 5.2%
- Ex-ante predicted volatility: 4.8%
- 1 year 95% Value-at-Risk: -7.0%
- Scenario 2008 Lehman Brothers default period: -15.4%
- Scenario Interest rates +100bps: +0.4%
- Scenario 2008-2009: -16.1%
- Scenario 2010 onwards: +158%