Leverage tends to get a bad rap in the retail market. In this post, we introduce a few common criticisms of leverage and explain why these criticisms are misguided.
Criticism #1: Leverage is dangerous.
Saying that leverage is dangerous is no different than saying that mean-variance optimization is dangerous. Leverage, like mean-variance optimization, is a tool that can be used to construct portfolios. Like any tool, leverage (and mean-variance optimization) can be dangerous when used improperly, but it is not dangerous in and of itself.
Holding a 10x levered position in SPY is a dangerous strategy that will almost certainly lead to ruin. Holding a 2x levered position in AGG produces a strategy with 11.1% volatility since September 2003 when AGG began trading. Over this same period, investment grade corporate bonds (LQD) had roughly the same volatility while McDonald’s (MCD) was almost twice as volatile. I suspect that most investors would not classify a diversified portfolio of investment grade corporate bonds or McDonald’s as dangerous investments.
Time for a pop quiz. Which of the two strategies in the following graph do you view as more risky?
Most readers probably said that Strategy 1 is more risky and this view is supported by a number of risk metrics. Strategy 1 has annualized volatility of 23.1% (10.8% for Strategy 2), a maximum drawdown of 55.2% (12.2% for Strategy 2) and an Ulcer Index of 17.7 (3.2 for Strategy 2).
So what are these mysterious strategies? Strategy 1 is an unlevered position in SPY. Strategy 2 is AGG with leverage/cash employed to target 10% annualized volatility (i.e. when AGG’s volatility is less than 10%, leverage is employed to increase the volatility to 10%; when AGG’s volatility is greater than 10%, cash is added to decrease the volatility to 10%). Over the backtest period, the leverage employed in Strategy 2 ranged from 0.5x to 4.0x.
Clearly, the danger of leverage can only be evaluated in the context of its use in a particular situation. When used intelligently, leverage is no more risky than unlevered, long-only positions in most standard asset classes.
Criticism #2: Leverage is expensive.
The main cost of using leverage is the interest rate charged on borrowed money in a margin account (or the equivalent cost of implement the trade through another financial instrument that can provide leverage like futures or swaps).
Interactive Brokers charges a blended margin rate of 71 basis points for an account with a balance of $5,000,000. From February 2004 to November 2012, AGG returned 5.0% with volatility of 5.8% for a Sharpe Ratio (assuming a 0.0% risk-free rate) of 0.86. Over the same period, the 10% target volatility AGG discussed above returns 10.4% with volatility of 10.7% for a Sharpe Ratio of 0.97 assuming borrowing costs of 71 basis points.
As this example indicates, leverage is not so expensive to employ that all benefits of using it are nullified.
Criticism #3: Levered Products Don’t Deliver On Their Stated Goals
The biggest culprits for this complaint are levered ETFs. There is misconception among many investors that levered ETFs are supposed to deliver a multiple of the underlying index’s return over any time period. The common belief is that if the return of an index for 2011 was 10.0%, the return of a 2x levered ETF on that index should be 20.0%.
When the returns of levered ETFs relative to their index are examined, it is easy to see that this relationship does not always hold. Consider the S&P 500 (SPY) and the ProShares 2x levered S&P 500 ETF (SSO). In 2011, SPY returned 1.9%. So SSO should have returned 3.8% (1.9% x 2), right? But when we look at SSO’s actual returns, we find that it was down 3.2% for 2011. Clearly, this levered product has not delivered on its promise!
Or has it? ProShares has this to say about SSO: “This Ultra ProShares ETF seeks a return that is 2x the return of an index or other benchmark (target) for a single day, as measured from one NAV calculation to the next. Due to the compounding of daily returns, ProShares’ returns over periods other than one day will likely differ in amount and possibly direction from target return for the same period.” ProShares explicitly acknowledges that situations like 2011 can and will occur.
SSO does much better when we compare its performance against its stated goal. We find that on 76.8% of trading days it delivered a return that was between 1.5x and 2.5x the daily return of SPY. This number would likely be even higher if we adjusted for the different fee structures of the two ETFs.
Limiting the investment universe to long-only strategies is akin to hunting with a bow and arrow: it may work in the appropriate conditions and if you have the requisite skills, but it’s a much more difficult long-term approach than using a rifle. The intelligent use of leverage can allow investors access to return streams that are robust to changes in economic environment, offer attractive downside risk protection and are cost-effective.