The Freakish Low-Beta Portfolio

by: Dale Roberts

It happened by accident largely, to be completely honest with you. Hey, that's how many things are invented. And the invention here is one of my accounts that delivers a beta that is so low it seems to defy logic and the laws of gravity. I achieved that mix from the process of selling some investments that had done extremely well from 2007 to 2011 -- and then the process of moving those assets to "safety." You can see an examination of the process and my recent investment history here.

It is certainly a safer portfolio that at one point offered an asset mix of roughly 68% bonds to 32% equities. That's a low-volatility mix of stocks and bonds for sure. In fact, using the very effective risk tools on Vanguard's site, that's ones of the safest mixes historically that one can construct when it comes to percentage of stocks to bonds.

Here's how that asset mix delivers on the volatility scale, using broad based indexes. It is certainly a broad and general evaluation, as it uses the broad based equity market and a broad based corporate bond index. From 1975, Vanguard used the Barclays Capital U.S. Aggregate Bond Index.

And for comparison, here's the annual price volatility of an all-equity portfolio for the same period.

There are obviously higher highs and lower lows from year to year with the all equity index. And historically, there are more gains to be had with that additional risk -- over time.

For the 68%-32% bond to stock asset allocation, it is rare for the mix to deliver a year with a 10% loss. There were zero events from 1980, whereas the all-equity model delivered 6 events of a 10% plus annual retracement that included two events of a 20% plus retracement, with one event delivering nearly a 40% drop in the last correction. Of course, the price swings (intra-year) for both models are more severe than the year-end figures charted above.

But my asset mix seems to go a step (or two) further beyond Vanguard's general guideline on volatility. When the market provided any modest correction over the last two years, my portfolio would only fall by about 20% of the total broad Canadian market retracement. Meaning of course, that when the markets fell by 10%, my portfolio value would only fall by 2%. And it would quickly move back to new highs. A beta of (0.2)? That's why I used the word freakish.

Beta is a measure of volatility in relation to the broader stock market. If an investment has a beta of 1, it is said to move with the market and deliver the exact volatility of that market. If an investment has a beta of 1.3, it is more volatile than the market, and vice-versa.

The portfolio mix offers some nice yield above 4.5% and has many holdings that offer low correlation -- and at times, some assets are negatively correlated.

For U.S. investors who shun volatility -- and those interested in portfolio volatility-reducing agents, I thought I would convert my Canadian low beta asset allocation to U.S. ETFs. There are two stocks in there, TransCanada (NYSE:TRP) and Enbridge (NYSE:ENB). I tried to get rid of them. I trimmed them a few times and moved the profits to the 1-5 year corporate bond ETF -- but those utilities kept growing back like weeds to modest positions. I get what they are trying to tell me. So I'll keep them. Maybe I'll trim them again soon -- they're already showing a double from the last trimming.

Here's the U.S. version of my Freakishly Low-Volatility Portfolio:

Bond ETFs Bonds - 68% Total Yield % of portfolio
1-5 yr Laddered Crp Bond (NASDAQ:VCSH) 2.1% (2012) 55%
Equity / ETFs Equities - 32% Total
Dividend High Yield (NYSEARCA:VYM) 2.9% 10%
Canadian Equity (NYSEARCA:EWC) 2.1% 5%
TransCanada (TRP) 3.7% 4%
Enbridge (ENB) 2.7% 4%

And here's the correlation matrix (over a shorter time horizon) according to coming out of the recession -- with robust equity markets, from 2009 to present.

And here's the correlation matrix of the portfolio during the 2007-2009 market correction and beginning of the recovery phase. Due to the short history of VCSH (November 2009 inception), I substituted the 1-5 year corporate bond with a 1-3 year corporate bond ETF (NYSEARCA:CIU). We see some more pronounced low, and even negative correlations through the market turmoil.

At the core of my low beta portfolio is a 1-5 year laddered corporate bond ETF. With concern for interest rates rising and the resulting decrease in bond fund prices that would result, I moved to one of the "safer" options in the bond world. It's on the short side, of course. The average duration is 2.5 years. The yield from 2012 is low as bond yields continue to be compressed. It will be lower for 2013. And in fact, the yield on the U.S. 1-5 year corp bond ETF is much lower than the yield on my Canadian bond ETF. That said, if interest rates begin to rise, VCSH will be replenishing with bonds that offer higher yields. The bond ETF's unit price will fall, but the yield should begin to rise.

From there, and to add some yield and diversification on the bond side, I added (okay, let's say sprinkled in) a high yield bond ETF and some longer dated bond ETFs. No one knows when interest rates will increase. Barbelling short to longer bonds can offer some lower correlation, with respect to the longer bonds to the equity holdings, and long bonds to shorter bonds.

On the equity side, I am certainly a yield pig in this portfolio. There are utility and REIT ETFs. That said, the U.S. versions of the ETFs offer a much lower yield than my Canadian holdings. The U.S. portfolio will deliver a yield closer to 3%, while my Canadian version has a yield above 4%.

One would think that this "safer" portfolio would give up a lot of potential total return due to its higher concentration in bonds. So let's have a look at how the holdings of this low-beta portfolio would have performed. The laddered corporate bond ETF VCSH is limited to November of 2009. This snapshot will demonstrate how this portfolio delivers in a healthy equity market environment.

The above returns in the Low-Beta portfolio asset allocation would have delivered an 8% average annual return from November of 2009. That's certainly well below the return of the U.S. equity index. Yet those returns are still quite surprisingly strong for a portfolio that offers only 32% exposure to the equity markets.

And how would have this asset mix performed through the great recession? As VCSH was limited to November of 2009, I replaced VCSH with the closest holding I could find (NYSEARCA:CSJ) -- Barclays 1-3 Year Credit Bond Fund.

And here's how that asset allocation would have held up from mid-2007 through mid-2009. According to a portfolio generating tool on CCN Money, the U.S. version of the low beta portfolio also offers a beta of (0.2).

The low-beta portfolio (green line) is already in positive territory in 2009, while the S&P 500 is still some 27% under water. The core volatility reducing agent is the use of a short-term bond fund in tandem with the broad based bond ETF. To be honest, the equity mix that I've included mostly serves to outperform the broad based market and offers a modest amount of volatility reduction through sector and international diversification.

For the above chart, I used the Dow 30 (NYSEARCA:DIA) due to charting limitations to represent the equity component. The equity mix in the low beta portfolio would have taken the portfolio into positive territory a little sooner.

The Aggressive Low-Beta Portfolio

What works in large amounts can also work in smaller amounts as well. If we flip the asset allocation (and tweak just slightly) and use 64% equities (in equal proportion -- each ETF getting just over 9% each) to a 36% bond allocation (bonds now also used in equal proportions), we can get some very nice returns and lower the volatility by a considerable amount. Here's how the aggressive low beta portfolio would have performed through the last market correction.

From a mid-2007 start point, the "aggressive" low beta portfolio outperformed the market by a considerable amount and limits the portfolio's downside by almost half. The bond mix is a very strong volatility reducing agent that can punch well above its weight, at least in the recent environment.

Conclusions and Final Thoughts

I'm not suggesting that you go out and recreate my low beta portfolio. Holding some ETFs and then throwing in two individual stocks is unconventional to say the least. That just happens to be what is in that account and I wanted to offer up the holdings in full -- even in its Frankensteinian state.

The main take away might be that you can consider using the combination of a short term bond ETF with a broad based bond ETF (plus high yield HYG and TLT) to lower the volatility of your portfolio -- whether you are an individual stock picker or indexer.

And please take note that heavy bond exposure presents its own risks. When interest rates begin to rise, bond fund prices will decrease. And perhaps the 32% equity allocation will be too low to buffer any price declines in the bond funds. And there's no guarantee that equities will soar when bond prices fall.

Due to rising equity markets, the addition of new funds and portfolio income that I've directed towards my equity ETFs, my low beta portfolio now offers an asset allocation that is closer to 60% bonds to 40% equities. I am certainly more comfortable to have a higher allocation to equities.

Managing risk is a moving target.

Disclosure: I am long DIA, ENB, TRP, VYM, EWC, EFA. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Additional disclosure: Dale Roberts aka cranky is a Streetwise Coach at ING Direct Mutual Funds. Streetwise Portfolios offer Canadians low-fee, complete, index-based portfolio options. Dale’s commentary does not constitute investment advice. The opinions and information should only be factored into an investor's overall opinion forming process