By Samuel Lee
Low-volatility stocks tend to be big, boring, and dividend-paying. HJ Heinz Company (HNZ) is a classic example. Boom or bust, Heinz ketchup sells with clockwork regularity, insulating Heinz's earnings from the business cycle. Stocks like Heinz are as bondlike as they can get. Interestingly, in nearly every market studied, low-volatility stocks have greatly outperformed high-volatility stocks on a risk-adjusted basis, a finding at odds with many investors' notions of risk and return.
There are three compelling reasons why this may be the case. The best is leverage aversion. Investors who target above-market returns may be unwilling or unable to use leverage to reach their expected-return targets. By resorting to volatile stocks (more accurately, high-beta stocks), which theoretically should outperform less-volatile stocks, they hope to earn above-average profits. Ironically, their collective bet on high-beta stocks leads to low risk-adjusted returns. Another is the fact that high-volatility stocks are systematically overpriced by investors seeking "lottery tickets"--stocks with a small chance of huge upside (Tesla Motors (TSLA), for example). Finally, asset managers tied to benchmarks and unable to employ leverage actually have a disincentive to own low-volatility stocks with below-average expected returns, regardless of how fabulous the expected risk-adjusted returns may be, and an incentive to own high-volatility stocks with above-benchmark expected returns, even if their prospective risk-adjusted returns are terrible.
The evidence behind low-volatility investing is truly impressive. However, low-volatility strategies can underperform during bull markets. iShares MSCI USA Minimum Volatility (USMV) can be expected to be about a third less volatile than the broad market. Owing to the strength of the evidence behind low-volatility strategies and a rock-bottom 0.15% expense ratio, this exchange-traded fund could serve as the nucleus of an equity allocation.
Depending on who you ask, U.S. stocks are either expensive or cheap. Using projected earnings, Wall Street analysts think the S&P 500 is cheap, trading at a forward price/earnings ratio of around 15, or a forward earnings yield just under 7%. That's not bad if you take it at face value. However, analysts are herd animals, and their projections often extrapolate the recent past into the future. When prices have risen, stocks become buys; when they've fallen, stocks become sells. This is an understandable outcome of analysts' short investment horizons and the career risk they run from being wrong and alone.
Value-oriented investment managers, who take the long view, have a more sober outlook. GMO co-founder Jeremy Grantham believes corporate profit margins are unsustainably high. Corporate profits as a share of gross domestic product is at a record 11%, well above the 6% historical average. The surge was in part driven by a massive expansion in public and private debt that accelerated in the 2000s, which has boosted aggregate demand even as the labor market has shrunk and wages have stagnated. In 1999, near the peak of the tech bubble, Warren Buffett wrote, "In my opinion, you have to be wildly optimistic to believe that corporate profits as a percent of GDP can, for any sustained period, hold much above 6%." If you believe 13 years is a "sustained period," he was wrong. Still, I think Buffett will be proved right eventually. One mechanism that may reverse above-trend profitability is higher taxes.
My favorite valuation metric, Shiller P/E, averages real earnings over 10 years to smooth out the effects of the business cycle. At the S&P 500's current price of $1,600, Shiller P/E is 23, about 40% higher than its historical average. According to AQR co-founder Cliff Asness, the stock market's average 10-year real return from this P/E ranged from negative 4.4% to 8.3% annualized, and averaged 0.9% annualized. The low average reflects the fact that Shiller P/E tends to be mean-reverting.
If you don't believe Shiller P/E will revert or expand, you can reasonably estimate the stock market's long-run real returns by summing current dividend yield and long-run real per-share dividend growth. The S&P 500 yields about 2% and historically has grown real per-share dividends by about 1.5% annualized. Add in net share repurchases, a hidden boost to yield, and investors can reasonably expect a long-run 4% annualized real return, about 40% lower than the 6.5% real returns the U.S. market has averaged since 1926.
USMV tracks the MSCI USA Minimum Volatility Index, which attempts to create the least-volatile portfolio possible out of U.S. stocks, under several constraints. The constraints revolve around trying to keep the index diversified on both stock and sector levels while capping turnover. If they didn't exist, the minimum variance algorithm may produce a very idiosyncratic portfolio. The constraints include keeping stock weights within 0.05%-1.5% of the portfolio, sector weightings within 5% of the market-weighted index, and a one-way turnover of 10%. The index is reconstituted semiannually. The index relies on the Barra Equity Model for two key outputs: estimates of risk factors for each security in the MSCI USA Index and estimates of how closely correlated each risk factor is to the risk factors of other securities. The outputs are used to estimate the "minimum variance" portfolio, the least-volatile portfolio of U.S. stocks possible. The model may help reduce inadvertent tilts beyond simple low-volatility equity exposure but introduces some uncertainty as to how the strategy will behave in the future. Despite this, we're confident the portfolio will be less volatile than its market-weighted parent index.
The fund's 0.15% expense ratio is market-beating in the nascent low-volatility category and very competitive in absolute terms. It's now at the point where it's about as liquid as its older, bigger competitor, PowerShares S&P 500 Low Volatility (SPLV). USMV's semiannual rebalancing and turnover constraints may keep its frictional costs below SPLV's turnover-indifferent quarterly rebalancing scheme. Like almost every iShares ETF, the fund engages in securities lending and returns 65% of the resulting revenue back to the fund's shareholders. The practice introduces some counterparty and collateral reinvestment risk.
PowerShares S&P 500 Low Volatility gathered plenty of assets soon after its launch. It's a fine alternative to USMV and our favorite low-volatility fund. It charges a decent 0.25% expense ratio. SPLV's index construction is admirably simple and transparent: Weight the 100 least-volatile S&P 500 stocks by the inverse of their volatilities, rinse and repeat every quarter. The simple methodology doesn't constrain the fund's turnover, or its stock and sector bets, so SPLV really piles on the defensive sectors. In back tests, both funds have about the same beta, or sensitivity to the market, of 0.70, but USMV's is a bit higher.
Investors can blanket the global equity market with low-volatility strategies using PowerShares S&P Emerging Markets Low Volatility (EELV) and PowerShares S&P International Developed Low Volatility (IDLV), which respectively levy 0.29% and 0.25% expense ratios. Both use SPLV's method for ranking and weighting stocks for their respective universes, so sector and country bets can be huge. The high turnover demanded by the strategy may be a problem for less-liquid emerging-markets stocks.
Investors worried about big sector and country bets should consider iShares MSCI EAFE Minimum Volatility (EFAV) and iShares MSCI Emerging Markets Minimum Volatility (EEMV), which charge compelling expense ratios of 0.20% and 0.25%, respectively. Oddly, iShares MSCI All Country World Minimum Volatility (ACWV) charges 0.34% of assets annually, more expensive than any combination of EFAV, EEMV, and USMV. While they may experience lower tracking error and turnover than PowerShares' ETFs, they sacrifice low-volatility purity.
Disclosure: Morningstar, Inc. licenses its indexes to institutions for a variety of reasons, including the creation of investment products and the benchmarking of existing products. When licensing indexes for the creation or benchmarking of investment products, Morningstar receives fees that are mainly based on fund assets under management. As of Sept. 30, 2012, AlphaPro Management, BlackRock Asset Management, First Asset, First Trust, Invesco, Merrill Lynch, Northern Trust, Nuveen, and Van Eck license one or more Morningstar indexes for this purpose. These investment products are not sponsored, issued, marketed, or sold by Morningstar. Morningstar does not make any representation regarding the advisability of investing in any investment product based on or benchmarked against a Morningstar index.