In a recent article, Mr. George Acs wrote about a pair of leveraged ETFs based on Silver bullion prices (see article here). The strategy Mr. Acs proposes is to purchase shares in each of Proshares Ultra Silver (AGQ) and Proshares Ultra Short Silver (ZSL). Once the shares are purchased, Mr. Acs than suggests selling calls with the shares as collateral (covered call positions). Theoretically, price movements in one should be cancelled out in the other. Therefore, the trader profits from volatility premium in the covered call, without betting on movements upwards or downwards in silver.
While I think this is a decent trading tactic, I tend to think it is a bit more risky than Mr. Acs indicates. Here are a few issues I see with the plan as it stands in Mr. Acs' article.
Tendency towards over exposure in portfolio
The expense for this strategy is significant for those of us with small portfolios. With AGQ selling at $39.50 and ZSL selling at $68.96 (as of closing May 31st), buying the minimum shares needed to exercise this tactic will cost us $10,846 before fees. If you are trying to limit your exposure to 10% or less, you would need a portfolio greater than $108,000. For the more conservative 5%, you will need $216,000.
Complicating taxes
Both of these ETFs are commodity based, and if held long enough will require a schedule K-1. This isn't such a big problem for many. But for those not familiar with K-1s (the forms IRS uses to handle income from master limited partnerships), understand your taxes are likely to get a bit more complicated.
Underlying assets aren't completely correlated
For this tactic to work, ZSL and AGQ must be inversely correlated. The closer the two are to being perfectly negatively correlated (-1.0), the safer the trade. Before trading this pair, or any pair, a careful study of correlation should be conducted. Here are two graphs (courtesy of TD Ameritrade Sinkorswim trading platform), using different timelines, demonstrating this pair's correlation. The correlation is displayed at the bottom of the graph, depicted as a red line. The lower the line, the closer to -1 correlation.
(Click to enlarge)
As you can see, in one week's time, the correlation between the long and short underlying ETFs varies between -1, and -.93ish. Not bad at all. The premiums from the option should cover the any negative effects we experience from the decaying correlation.
This image tells a different story than the first. Over longer periods of time, correlation can decay significantly. At one point in July, it appears that correlation slipped to -.3, which means this position was effectively directional. If you were on the wrong side in July, one side of your position could have lost significant money, and the other side, would have gained very little. Your premiums may not have covered your losses.
One of the reasons these positions have correlation decay is because of the opposing forces that compounding returns have on these leveraged positions. Here's a graph that demonstrates over ten trading days, how returns could slip from where they were designed to be.
Silver | AGQ | ZSL | |
Day 1 | -1.10% | -2.20% | 2.20% |
Day 2 | 1.30% | 2.60% | -2.60% |
Day 3 | -2.30% | -4.60% | 4.60% |
Day 4 | -1.60% | -3.20% | 3.20% |
Day 5 | 1.40% | 2.80% | -2.80% |
Day 6 | -1.50% | -3.00% | 3.00% |
Day 7 | 1.80% | 3.60% | -3.60% |
Day 8 | 3.50% | 7.00% | -7.00% |
Day 9 | -2.50% | -5.00% | 5.00% |
Day 10 | -1.40% | -2.80% | 2.80% |
total return | 97.44% | 94.58% | 104.10% |
ideal return | 97.44% | 94.88% | 1.0512% |
In this case, your leveraged long position would have lost more money than 2 X Silver, and your leveraged short position would have gained less than 2 X inverse Silver. Keep these positions open long enough, and significant deviations can occur.
Alternate strategy #1
Option strategies allow us to control the same position with less money. Additionally, holding options on the underlying position is not the same as receiving income from master limited partnerships. Of course, now we are effectively leveraging a leveraged position, so we will need to be very careful with execution, as well as continually watch our positions.
My first thought would be to open a calendar spread on both ZSL and AGQ. A calendar spread is opened by purchasing one option at X strike price with an expiration date sometime in the future. Next, I sell an option of the same type and same strike price, but with an expiration date closer to the current date. I am trying to gain from both time decay difference between the two positions, and any difference in implied volatility in the two positions. Here is an example:
Buy 1 X ZSL 65.00 Jan 2013 Put contract for $1570.00. Sell 1X ZSL 65.00 July 2012 Put contract for $665.00.
Buy 1 X AGQ 41.00 Jan 2013 Put contract for $950. Sell 1X AGQ 41.00 July 2012 Put contract for $410.00.
The advantage to this trade over owning the shares outright is that I control 100 shares of each ETF for an initial outlay of $2520, rather than $10,846. I am immediately credited with $1075 on the outset of the trade. However, I still don't know what my final return is going to be. Since these puts are both just out of the money, and I expect one underlying asset to go down when the other goes up, if the price of silver bullion is either higher or lower than it is today, I know one of our short puts will be in the money on July 20th (expiration date of the short puts). I will need to buy back the "in the money" put on that date, which could eat up a significant portion of the premium I earned at the beginning of the trade. Of course, the assumption is that everything I lose on the short put, will be earned on the long put as intrinsic value.
The bigger concern I have is that if implied volatility falls on the long put positions, the value of those positions could fall drastically despite price movements one way or another. Because the MAXIMUM amount of premium I will earn between now and July 20th is only 43% of our total investment, I need the long put positions to retain significant value. If they both lose value due to a fall in implied volatility, I could easily find myself on the losing end of this trade.
A final concern lies in the fact that a calendar spread makes most of its money if the underlying ETFs do not end up with gains or losses on the expiration date of the short put. Ideally, I want to keep the implied volatility, but end up with the share prices in the same place as the start of the trade. The odds for maximum gain are not in my favor.
Alternate trade #2
I can nullify some of my concerns by buying long puts at the farther out expiration that are in the money, and selling out of the money puts with nearer expirations. By doing so, I can reduce (somewhat) the deleterious effects that a reduction in volatility will have on my long position (especially if they are deep-in-the money), and give each spread a larger profit range, because the option can profit if the underlying asset rises slightly, stays stagnant, or falls (as long as it doesn't fall too drastically). When combined with an identical position negatively correlated, this should result in a position where I stand to profit under a variety of silver price movements. If prices end up in the same spot, I will make money on both the ZSL and AGQ spreads. If the price of silver moves up a little, I could potentially make money on both positions as well, and the same goes for a gradual price move down. In the case that prices move sharply up or down, the position should be hedged to some degree with one side making money and the other losing money. Whether the final outcome is a profit or a loss will depend entirely on what strike prices I chose for all four positions of the trade, and how much premium I earned at the start of the trade.
Ordinarily, when I open a diagonal spread, I like to purchase the put (or call) as a Deep-in-the money LEAP. Deep-in-the money options tend to act like the underlying equity. When the stock gains 5% , the option will gain around 5%. Both ZSL and AGQ have Jan 2014 puts, but they aren't very liquid, so I'm going to pay the price for using them. Buying a $50 ZSL put and a $30 AGQ put will cost me around $5,000, but with slippage I start the position down several hundred dollars.
Additionally, I usually use this position to allow me to maintain a position for a 12 to 18 months. This timeline gives me a broader timeline to prove correct on directional bets. In this case, I'm deliberately avoiding a directional bet, and, as I demonstrated earlier, longer timelines in this trade are a source of greater risk, in the form of decaying correlation.
Therefore, a diagonal spread with nearer expiration dates may provide the best compromise. First, the long position will provides me with more profit potential than the horizontal spread. Next, by not buying the LEAP, I reduce my total commitment and raise the percentage the short position earns me. Additionally, nearer expiration dates are more likely to have more volume, so I can reduce the slippage. Finally, the shorter duration on the long position forces me to close the position before the correlation between the two positions decays to significantly.
Here is an example of the spread (an overview of the diagonal spread strategy can be found here:
Buy 1 Aug 18, 2012 ZSL 69.0 put for $1390.00 (bid ask spread is 13.10-14.50, which is by far the worst of the group. All others are within .50).
Sell 1 July 21, 2012 ZSL 54.0 put for $260.00
Buy 1 Sept. 22, 2012 AGQ 52.0 put for $1190
Sell 1 July 21, 2012 AGQ 39.0 put for $190
A trade with these numbers should give me $450 on an initial investment of $2580 before fees, or 17.44%. I set the longs to be around 15% in the money, and shorts to be around 10% out of the money. Greater buffers on both ends might be prudent, since these are leveraged ETFs, and price swings are bound to be extreme. If I'm extremely lucky, and all short positions close out of the money on July 21st, I can always write another short position for August.
NOTE: I would prefer to keep strike prices the same for ZSL and AGQ, but they do not have any matching strike months between July and January. The August long put will lose time value faster than the September long put, especially if I write another short put at the end of July. In this particular case, it would probably be wise to wind the position down in July, no matter what the outcome is.
Conclusion:
Mr. Acs wrote an intriguing article a week ago, marrying two ETF positions and trying to benefit from their inherent volatility. While seemingly simple to execute, trading these two positions with the ETFs hedging short calls would be a huge percent of most small time trader's portfolio, and holding the positions for any length of time over a month exposes the position to risk due to potential decreasing correlation. Trading long options, while significantly increasing the complexity of the trade, could mitigate some of the dangers, boost returns, and allow a smaller commitment of capital to the overall trade. For those willing to brave the seemingly frightening and bewildering world of option spreads, these pairs may prove to be very profitable.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
