Arbitrage Experts Feast On The Public's Ignorance

by: Peter F. Way, CFA


Leveraged ETFs come in two flavors, sweet and sour, long and short (inverse).

Uninformed reasoning often holds that they should produce equal and opposite coming price changes when based on the same underlying equity market indexes.

Actual results are so different that legal warnings have had to be issued that further confuse and misdirect the investing public.

Can the resulting misperceptions offer further magnificent profit opportunities for investor/speculators with adequate understanding of what is really going on?

An astute reader of my just-published SA article on Market-Maker [MM] outlooks for coming market direction raised the question of why the price behaviors of leveraged long and leveraged short (inverse) ETFs are not symmetrically opposite one another. Good reasons exist for the disparities, and will be explored here.

What's really going on?

Sophisticated market arbitrageurs, whose principal parallel occupations encourage their persistent presence, simply enjoy a daily harvest in this financial Garden of Eden, plucking the value of investment time as it ripens on these short-structured inverse ETFs, continually reducing the ETFs' market quotes. For most of those ETFs the effect has been horrendous to any buy-and-holders of the inverse leveraged items.

(source: Yahoo Finance)

The effects creating these treats (for shorters of the inverse ETFs) are variously referred to as contango, roll costs, and theta, depending on the descriptive environment at hand. They are the principal reason for the lawyer-warnings that ANY leveraged ETFs should be held for one day only. Safe advice for the lawyers, but grossly wrong for the other side of the trade. That bad advice for those contemplating buying the leveraged long ETFs is clear in this picture of the progress of the siblings of the above-pictured evil twins.

(source: Yahoo Finance)

These longs have benefited mostly from the rising market, identified by the yellow S&P 500 Index line in both pictures. But while the S&P rose 100% during the 5+ years, and the 3x longs rose 300% to 500%, the shorts declined variously to only 6% to 2% of their initial cost values.

Investors should look at how these long-leveraged ETFs' prices rise across time, and understand that they can be used as longer-term investment vehicles, not solely as one-day trading tools. The problems come with the inverse leverage ETFs. They are not suitable for long-term ownership, only for long-term shorting as a speculation.

The specifics of these pictures are revealed in the following table, which starts off showing how the indexes being tracked by the ETFs were behaving during the ETFs existence.

The dates used here have been dictated by the existence of the ETFs' first availability to be traded and to be able to have their expected coming prices determined. Excluding the two bottom ETFs, the others track the indexes indicated at the top of the table. More about the bottom two later.

The first five ETFs are all 3x leveraged longs. Over the course of up to 5 years, they all grew by about 200% to over 600%, what some have colorfully referred to as 3-baggers to 7-baggers. Ah! The power of compounding, and the value of TIME.

The second five ETFs are the evil twins, the inverse, or short-structured ones.

The indexes during this period all were in a largely steady rise, so the longs gains should be no surprise. A 3x leverage on an 18% gain rate should produce a +54% rate. It also produces, at the inverse, a -54% rate. The surprise comes at what is left when these rates are applied to a common $1.00 base.

On the plus (or long) side, +54% a year turns the $1 into $7.75, or so. On the minus (or short) side it turns the $1 into three pennies, or so.

Hey, they ain't no free lunch, market-wide. When somebody makes it, somebody pays for it. (This may be where the tort lawyers got their bright, destructive ideas.)

Why are the results of ownership not symmetrically opposite?

What may be hard to understand, and appreciate, is just how finely the payments mechanisms work. Blame it all on derivatives (as the politicians like to do). They are the basic tools of arbitrageurs. And arbitrageurs are the price police of the marketplace.

Just as perpetually hungry seagulls keep the beach clean, arbitrageurs snap up every risk-free opportunity that an otherwise careless or sloppy market participant (including professionals) puts out for grabs. Often their opportunities exist because the transmission lines are fairly complicated, involving sometimes apparently tiny message packets, in places not often frequented or adequately understood.

Take leveraged ETFs. In order to construct these instruments so that their prices will move at multiples of the underlying index they track, derivatives called financial futures are involved. Such futures provide a bet on what the coming price of a financial instrument will be sometime in the future. A bet that the price will be higher at the expiration date of the bet, more than it costs now, makes you money if you're right, costs you money if you're wrong. A legal contract makes sure that what should take place on the expiration date actually does happen.

Where does the leverage come from?

The neat attraction to futures is that while the contract obligates the parties to an outcome, there is no stakeholder as such to the bulk of the value involved in the contract. Instead, it is, in fact, done with "borrowed money." So financial leverage is available to accentuate the price moves of the underlying subject of the bet.

Since the leveraged ETF bets are based on stocks or other securities, largely, the margin rules of the SEC must be observed. For decades, the margin allowable on stocks has been 50% - the investor can borrow half of the value of the shares in question, a 2 for 1 leverage. The ETF engineers figured out a way to put another 50% margin on the ETFs, so instead of putting up 50% of its cost they need put up 50% of the 50%, or only 25%, a 3 for 1 leverage.

So instead of a 2x instrument, a 3x bang for the buck can be had. A way to get rich quicker, or to go broke quicker.

But in the process of getting any ETF leverage, 2x or 3x, doing it with stocks is not desirable, for several reasons, some of which are arcane enough to do most of us no good to try to explain. The answer is to use futures contracts to provide the degree of leverage desired and that introduces a dimension overlooked by many investors - TIME.

If you hadn't noticed, TIME is a four-letter word. Other four-letter investment words are RISK and LUCK.

These futures contracts, that provide the ETF leverage, deal in both price and time. The time component is usually sold in monthly increments for a settlement price, like a FEB26, where the buyer at the expiration date (in February) gets to "sell" his contract to the buyer, who is obligated to pay the settlement price (of $26).

The open market for financial futures sets its everyday prices by open auction bidding, similar to the way prices of other financial instruments are negotiated. Expectations are a part of the scene. If the subject is expected to rise in price between now and the expiration date, monthly futures contracts further out in time will likely be priced at progressively higher levels.

The ETF's promise is to move in its specified direction, relative to the market's move - the market increases +1% in a day, the 3x leveraged long rises +3%, The market declines -1%, then the ETF may change by other than -3%. Why? Look at the table above and see what the total % change numbers are for a +50% rate and a -50% rate when magnified by time, as in SPXL and SPXS. The effects are far from equal.

Keeping the leverage consistent as prices change

In order to make tomorrow's one-day ETF % change equal to today's one-day % change in terms of the market change, the proportions of leverage in the ETF's futures holdings need to be changed a bit. So every day there is a little change going on by the ETF manager, buying some expiration date futures and selling others. Since they are not likely to be priced equally, one side of each trade is going to benefit and the other side is going to incur a cost. The way that works out usually is to the benefit of the longs, and to the cost of the shorts.

To see the uneven impact between Leveraged Long and Leveraged Short ETFs, take a look at how their actual prices have progressed for 3x Dow-Jones tracking issues, ProShares UltraPro Dow 30 (NYSEARCA:UDOW) (the long) and ProShares UltraPro Short Dow 30 (NYSEARCA:SDOW) (the short). We chose these because they are among the least extreme and are representative of the daily recalculations effect on most leveraged ETFs. Additionally, we have alternating periods of time where the long side and the short side is in favor and the recalculation impacts can be seen cumulatively through time.

Note the two periods where the DJIA index (blue line) declines and SDOW, the -3x ETF, rises in price appropriately, in May through July, 2010, and July through October, 2011. Otherwise, SDOW's price (orange dots) is mainly just a wiggling decline through the three years.

An exactly parallel in time picture of UDOW, the +3x DJIA-tracking ETF is below.

As it should in a rising market, UDOW climbs aggressively except for the two periods noted for SDOW's limited moments of glory. But what is important here is the converging tracks of the ETF price (orange dots) and the Index price (blue line).

We have arranged the scale of the ETF (left side) with the scale of the index (right side) so that they are exactly proportional. If the ETF simply benefited from the leverage, the two tracks would be parallel, rising equally.

Instead, they converge, with the advantage going to a faster rise in the ETF. This is due to the advantage of resetting the hedge ratios of the leveraged ETFs daily so that they will produce on a daily basis the 3x relationship promised.

Market-wide the advantage gained by UDOW in the process has to be made up somewhere. It comes right out of the prices of SDOW. This effect is present in every pair of leveraged ETFs, with the advantage usually going to the long-leveraged twin, at the expense of the inverse ETF.

Now about the other ETF twins

ProShares Ultra VIX Short-Term Fut ETF (NYSEARCA:UVXY) and ProShares Short VIX Short-Term Fut ETF (NYSEARCA:SVXY) are very different ETF animals because what they track tends to behave quite differently than market indexes, and often acts the opposite of equity indexes. They track market uncertainty, as represented by option premium spreads between contracts.

Conventional option pricing tends to follow standard formulas involving the market price of the subject underlying the option, the strike price of the option contract, the length of time until the option expires, interest rates, and the potential perceived risk involved in undertaking a position in the option at this point in time. They all can be related via a formula to determine a "standard" price for any of the usually several option contracts for an underlier at various strike prices and times to expiration.

But instead of solving the formula to find prices for the options, if the current market prices for the options are taken as inputs to the formula and the equation is turned around, it can be solved to determine the degree of uncertainty, or implied risk, that is seen in the price of the underlier. The result of this process is usually referred to as "implied volatility." UVXY directly tracks the index of volatility [VIX], and SVXY tracks its inverse value changes.

But here's the twist. When equity prices drop and investors get scared, implied volatilities rise. Got it? Stock prices down, volatility up. When things return to "normal" and stock prices recover, volatility and the VIX go back down. Which is where the VIX usually lives most of the time.

So the behavior of these ETFs is just the opposite of the others. The inverse SVXY usually travels in parallel with market prices over time, and the direct UVXY plays the opposite role, usually the assignment of the inverse ETF.

Looking at the table, it can be seen that even in their short lives (less than 3 years) these ETFs have had major moves, at annual rates outdoing most leveraged ETFs. The price volatility induced into these ETFs just by the market is for most investor/speculators plenty by itself; they don't need any leverage, but they have 2x leverage.

Because there is no tradable equity for implied volatility, these ETFs must resort to holding futures instruments on the Volatility Index. The futures have finite lives, and must be periodically moved forward (or "rolled" in the trade vernacular) so they encounter some of the same effects of that trading activity as the leveraged ETFs.

And since everything is turned around in this mirrored world, instead of the inverse instrument getting hurt, it gets helped. The history to date, as indicated in the table above, shows how powerful a simple buy and hold on SVXY has been. The exception proves the rule.

Of course, with productive timing information, even these outrageous annual rates of gain can be improved upon. But beware the ultra-toxic nature of a long position in UVXY.


Leveraged long ETFs can be held productively and with very favorable risk~reward tradeoffs, while short-structured, or inverse ETFs are indeed dangerous instruments in any applications other than very short-time holdings.

The special convoluted nature of the inverse VIX ETF SVXY endows it with attractions not found elsewhere and should be considered as a long-term holding, where appropriate, in any wealth-building equity portfolio.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. 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.