From the October 2008 issue of Wealth Manager Web • Subscribe!

October 1, 2008

VOLATILITY SURPRISES

Volatility can be an investor's best friend, but friend can turn foe at times. The first line of defense is understanding volatility's role in money management and being prepared for the Jekyll and Hyde possibilities. That's good advice for investing generally, although it's essential when using one or more of the increasingly popular levered exchange-listed funds and their mutual fund cousins.

Common sense, right? Perhaps, although it's easy to get confused when it comes to the finer points of volatility and how it affects returns. The basic relationship between volatility and performance is widely understood--higher returns generally come at the price of higher standard deviations. The linkage is more complicated once leverage is added to the mix. Yes, everyone knows that leverage elevates volatility, but that's only part of the story.

The critical factor with levered funds is the interim level of volatility relative to return, as outlined in a recent research report by Mariana Bush, ETF analyst at Wachovia Securities in Washington, D.C. "We are concerned that too many investors are not sufficiently aware of the implications of the volatility of the underlying index on the return of the leveraged product," she wrote in a July 2 briefing to clients.

Part of the problem is that the basic intuition about volatility and returns can be misleading. As Bush details in her research, when volatility rises above a certain point in a levered index fund, a positive return in the underlying target benchmark turns into a lesser performance or even a loss for the portfolio.

To illustrate how leverage can reshuffle returns in unexpected ways, Bush lays out three examples (see Table at end). In each of the three volatility scenarios, two series of hypothetical net asset values are compared over a five-day period. One is for an unlevered index (1x beta); the other represents a 200-percent levered fund (2x beta). A third column shows the daily percentage change for the unlevered index (Daily Return). Keep in mind that in all three scenarios, the unlevered benchmark (1x beta) posts a 5 percent return at the end of the five days.

As Bush's analysis shows, leverage can introduce a wide array of returns that differ from the performance of the target index. No surprise there, but here's where it gets tricky: The fact that the underlying benchmark rises over a given period doesn't guarantee a gain in a levered fund intent on magnifying the benchmark's performance. Consider Scenario 1, when the 2x levered fund suffered a small loss even though the benchmark rose 5 percent. Meanwhile, in Scenario 3, the levered fund more than doubled the benchmark's 5-percent gain. Again, the benchmark return is identical in both cases and yet the levered funds report radically different results. Why? Daily volatility levels were different. The fact that the end return for the underlying index was the same in both cases was irrelevant for determining how each levered portfolio fared.

That leads to the fundamental rule that dictates performance in levered index funds compared with the target benchmark: Interim volatility matters--a lot, in fact, and quite possibly far more than is generally recognized. The lesson applies to levered inverse funds as well as long levered funds.

Exactly how much volatility matters is closely tied to how frequently a levered fund rebalances and how much leverage is employed. The more frequently the rebalancing takes place, the bigger the volatility-related impact on a levered fund's returns. All else being equal, the effects of volatility are more potent in a fund that rebalances daily (which is typical for most of the assets under management in levered exchange-listed portfolios) compared with a monthly rebalancing program. The ideal scenario for a levered fund occurs when volatility is calm relative to return.

The basic message: Developing expectations for both volatility and performance is critical with levered funds. Yet--as Bush notes in a recent interview with Wealth Manager--it's not obvious that analyzing the path for prospective standard deviation is widely practiced among investors using levered funds. That's an ominous thought if you consider that the use of levered exchange-listed securities has risen sharply since the SEC gave the products a green light several years ago.

Despite the risks, Bush emphasizes that levered funds can still be valuable portfolio tools. Assuming, of course, that the investor is conversant in volatility and how it can play havoc at times with otherwise prudent return expectations.

What is the main risk with levered funds?

The biggest risk is that you don't get the return that you think you're going to get. Even if you correctly predict the return of the underlying index, you can be disappointed because the fund does something completely different and delivers a much lower return than you thought it would.

Leverage is the issue, whether you go plus two times or negative two times. You're magnifying the moves, the volatility. But you don't get the same [volatility distorting] effect on performance with a straight inverse fund. If it's a one-time inverse, it's the same as the plus one-time beta, but in reverse.

Let's say that you forecast a 10 percent return for an index over the next 12 months. Feeling confident, you buy a 2x levered ETF targeting said index. A year later, your index forecast is proven correct. But you still may not get 20 percent, or two- times the index return?

That's the shocking part: Let's say the underlying index does 10 percent. In that case, nobody thinks of the possibility that you could lose money by owning a levered fund targeting that index. Nobody thinks that the return on the levered fund could be a negative 5 percent. But if volatility is too high relative to return, that's possible.

According to your report, the more frequently a levered fund is rebalanced, the higher the potential for delivering returns that differ from the unlevered index. How often are the listed levered ETFs rebalanced?

Almost all the funds are rebalanced on a daily basis. One thing that I learned after I wrote the report, however, is that the Deutsche Bank leveraged funds rebalance monthly, not daily. I assumed that all of the levered funds rebalanced daily in the report. ProShares rebalances funds daily, and ProShares has 98 percent of the assets [in levered and inverse exchange traded products]. But there are a few that rebalance monthly, and if it's monthly, the effect of volatility shouldn't be as strong compared with daily rebalancing.

On the other extreme, assume that you never rebalance--meaning that volatility should not be an issue [in terms of changing the target index's return over an investment horizon]. You could go into a structured product, such as a swap, for instance. I'm sure someone can produce a security that says that it will deliver exactly two-times an index in say, 30 days from now. But we're not dealing with those products with exchange-listed funds or mutual funds.

How should investors think about volatility and levered ETFs?

Looking at historical volatility for the underlying index is a good start. But you need to focus on expected volatility, too. Looking at expected return isn't enough--everybody has an idea about that. You should at least be saying something like, "The volatility has been X; now let's stress test it." What if we get 25 percent or 50 percent extra volatility? In that case, you'll have a feel for what higher volatility will bring.

Is thinking about volatility in this way, including forecasting it and studying its record, widely practiced?

A lot of people don't understand volatility. I've talked with several investors--institutional investors--and I haven't found a single one that told me that "We account for volatility; we monitor it." After I talk to them about it, they tell me, "Now I understand why [the levered fund] isn't working as well as it should." You don't hear too much about volatility expectations. Maybe it is something that will be developed much more, and it probably should be. Generally, the people who understand this issue are the people who understand options and managed futures. Why? Because they deal with volatility in those products all the time. There's a lot of leverage in managed futures.

Levered funds are said to be useful for portable alpha strategies. A basic example is a portfolio targeting a 50 percent equity weighting. Alternatively, using a two-times levered fund for the equity allocation, the actual weighting could be reduced by half, to 25 percent, thereby leaving assets available to invest elsewhere while still capturing the beta effect of the original 50 percent equity strategy. Are levered exchange-traded securities good tools for such strategies?

I'm not sure that's going to give you the return you expect. Maybe you're lucky, and volatility is close to zero, but I wouldn't count on that. The only way that it would be sure to work is if you rebalance every single day.

Or, if the levered fund had a one-time end of period rebalancing?

Yes, but that wouldn't apply to any of these exchange traded products. In that case, you'd be using a structured product--a swap or something else.

Once again, the point is that if interim volatility is too high, the investor probably won't get an exact multiple of the return from the underlying index.

Yes, volatility is a danger. These are great products--there's nothing wrong with them. But you've got learn how to use them.

What is the ideal use for levered ETFs and the like?

When you want a very strong bet on something, or if you want to hedge a position. For example, if you were very negative on a sector, you could use an inverse sector fund or a levered inverse sector fund, depending on the confidence in your forecast. But you have to be careful: If volatility goes up too much [relative to return], you're not going to earn what you think you'll earn, even if you're right on the bet [about the underlying index].

Does that mean that levered ETFs are better suited for tactical rather than strategic decisions?

I'd use them as a tactical product, as opposed to a core holding.

So the best-case scenario for a levered fund is an index return that meets expectations in a period when the benchmark's return volatility is relatively low.

Exactly. That's your perfect scenario [for using levered ETFs]. Volatility is close to zero. By zero volatility I mean that returns are identical each day.

James Picerno (jpicerno@highlinemedia.com) is senior writer at Wealth Manager.

Reprints Discuss this story
This is where the comments go.