The continued trickle of better than expected US macro data along with a welcome decline in fear out of European financial markets has led to quite a risk rally since the beginning of the year with the S&P 500 (SPY), IVV, VOO up 12% and we're not even through the first quarter yet! With investors seemingly feeling more confident, money has been flowing back into risky assets, perhaps a bit too enthusiastically given that there's still the small matter of trillions in debt deleveraging to put governments and individual households on sound long-term financial footing.
With this as backdrop, I thought it would be a good opportunity to revisit a foundational principle of good long-term asset allocation: lowering risk through diversification. As the saying goes, there may be no free lunch in investing, since to get a return, you're taking on some form of risk(s) but diversification (i.e. mixing different asset classes that have low correlation) has historically reduced overall portfolio risk.
Before I get a lot of howls on how diversification failed during the financial crisis, let's clear up a few things: First, assembling a diversified portfolio through combining low correlation assets was never intended to eliminate risk completely. Second, long-term investing means just that, long-term. If we're evaluating how well diversification held up during the recent financial crisis for the average investor's retirement portfolio, let's look back at least 10 years and ideally through 20 and 30 year time periods.
Notwithstanding the brutal equity market gyrations of the past 10 years, an investor holding the 60/40 Vanguard Balanced Index Fund (VBINX) has seen an average annual return of 5.6% for the past 10 years. Certainly not a disaster. In fact, a great reminder of how even one of the simplest asset allocations can diversify equity risk well even during one of the most tumultuous periods of investing we've seen (and hope to never see again) in our lifetimes.
By far the most common and well accepted form of diversifying equity risk is through bonds. Historically U.S. government bonds (SHY), IEF, TLT have been an effective source of diversification for equity risk, which intuitively makes sense particularly for shorter-term bonds which we think of as almost the exact opposite of a risky asset. In fact, if you look at the holdings of the Vanguard Balanced Index fund I mentioned above, you'll see that of its 40% bond allocation, just over 70% is in US government bonds, part of the reason for its relatively strong overall performance the past decade.
Given the unique set of circumstances though that have accompanied the government bond rally the past decade where IT Treasuries have had a 7% average annual return, it's fair to wonder just how well Treasuries can diversify equity risk over the next ten years. Most starting yields are now below 2%, meaning expected real returns on most Treasuries are close to zero if not negative. No doubt Treasuries can rally if there is another financial or other global crisis, as there will always be a short-term flight to Treasuries for their safety and liquidity, which at current ultra-low yields is probably the strongest investment rationale for maintaining some weight to Treasuries in a core fixed income allocation.
What's the alternative? Individual investors have cost effective access to many more types of bonds now than ever before. Though historically Treasuries have provided a superior level of diversification, looking forward, I would balance that with fixed income investments more likely to produce at least a small amount of future positive real returns, so high quality corporate bonds (VCIT), LQD and hedged international bonds (EMB) all deserve a look for enhancing a core fixed income portfolio for the next decade.
Gold has been a financial asset and even money itself for thousands of years. Its nearly indestructible nature, limited supply and lack of other commercial uses has allowed it to continue in its unique role, albeit in a much more limited (some might say rather arcane) capacity in our modern financial system.
Let's put aside the endless debate though on the nature of gold as money or metal and focus instead on its impact on portfolio diversification as a distinct asset class. During the past nearly 40 years when gold prices have been decoupled from the dollar and other currencies, gold has had a low to negative correlation with equity market returns, adding a source of diversification.
Also like Treasuries though, gold has had quite a run the past decade. The advent of several popular and liquid gold ETFs (GLD), IAU has made it much simpler and less costly for investors to add gold to their portfolios. Given that gold has risen nearly 150% during the last five years while the S&P 500 was basically flat, we're faced now with the same question to ponder as with Treasuries: Given the unusually strong performance we've seen the past ten years, what should we expect for the next decade? There's no consensus on how to value gold, let alone its future expected returns, but I think it's fair to assume that the next 10 years are not likely to be a repeat performance of the past decade.
I'm also a little more concerned about gold than Treasuries when it comes to diversification because the longer-term negative correlation I mentioned above for gold has turned more positive the past year, as you can see in this correlation matrix below, indicating that gold is now moving more in tandem with stocks.
1-Year Correlation Matrix ending 2/29/12
iPath Exchange Traded Notes S&P 500 VIX
iShares Gold Trust
iShares Barclays 7-10 Year Treasury Bond Fund
iShares S&P 500 Index Fund
I purposely saved VIX (the "fear gauge") for last, because unlike bonds or gold, holding VIX exposure long-term is a good way to lose money in a portfolio. Without getting into all the nuances of futures contracts, there's a basic concept to keep in mind. Unlike a cash bond or a physical commodity like gold that you can buy and literally hold, volatility is a measure that can only be traded by some form of contract or derivative -- essentially two parties agreeing to be on either side of a view about the measure of future volatility. Sound wonky? Hey, if you're one of the happy Vanguard Balanced Index Fund investors above, please stop reading now.
Seriously, please don't get confused on this point. Even the ETF and ETN versions of VIX (VIXY), VXX are giving you either a wrapper around futures contracts or in the case of the ETN like VXX, a debt instrument offering you the return of the S&P 500 VIX short-term futures index, net of fees.
In other words, a view on VIX is a bet on the future volatility of the S&P 500, not a long-term investment. With that said, you can see from from the correlation table above that VXX has had a high negative correlation to the S&P 500 the past 12 months (3-year correlation is similar at -0.7), so it can be an attractive short-term hedge or tactical bet particularly when you think investors are growing overly enthusiastic or complacent about downside risks. Unlike gold and Treasuries, it also has one other advantage right now: it looks relatively inexpensive as it's fallen to a 5-year low while most other asset classes have rallied strongly.
For long-term investors, diversifying equity risk has become more challenging as risk assets have risen substantially from crisis lows, while at the same time, bond yields are at or near historic low levels. Investors should look to high quality corporate and hedged international bonds to enhance their fixed income portfolios.
Gold has also rallied significantly the past five years, so its future ability to diversify equity risk will likely be muted, though gold may have some unique benefits as a future inflation hedge.
Adding VIX exposure through ETFs, futures or options is not a solution for long-term portfolio diversification, but may be a good option for experienced investors looking for a short-term hedge to the unusually strong equity returns we've seen so far this year.