I’ve been playing around with ETFs for a while and thought about a way to generate a considerable profit from them. Option traders are familiar with time decay but it also applies to leveraged ETFs. This ETF time decay is a result of the fund management fees, leverage financing interests and other expenses. Since November 2008, there are 2 interesting funds for this strategy: The Direxion Financial 3X ETFs. FAS is long the financial market and FAZ is short the financial market, both using 300% leverage. These ETFs are very popular and are traded in the tens of millions shares a day which make them good candidates for short selling.
While I was looking around for information on this strategy, some people rightfully said “they cannot go both to 0”. This may be true but that’s not the object of this strategy. Here, we want to re-hedge the positions daily to balance the dollar amount on FAS and FAZ. This way, we profit from the downward bias on both funds and we are hedged against the market since the net delta of this position is near 0 and readjusted to 0 every day.
The strategy is: everyday, at close time, balance the positions on FAS and FAZ to have the same short dollar amount resulting in a net delta of 0. A quick backtest starting with a short position of $5000 on each ETF shows:
| Start amount | $ 10,000.00 |
| Gross profit | $ 2,975.22 |
| Fees ($0.005/share) | $ 376.31 |
| Days | 125 |
| Annual return % | 87% |
| Annual return | $ 8,687.64 |
| Average win $ | $ 107.06 |
| Average loss $ | $ (60.10) |
| Wins | 48 |
| Losses | 36 |
| Win % | 57% |
Note: I haven’t tested this strategy. If you are aware of any reason why it would not work, please let me know. As of this writing, the short availability for FAS and FAZ is around 400,000 shares each with my broker.
Update: I fixed the annual return percentage and amount calculations. The 126% figure was wrong. It actually is 87%.
Update 2: As pointed out by AJG, watch out for dividends. As a reminder, when you are short, you have to pay the dividend. Usually the stock will drop by the same amount as the dividend so this isn’t a big problem.
This post has gotten my attention. I've been going over your results on my spreadsheet and have similar results. Interestingly, if you decide to rebalance only on the days that your account balance sets a new high, your return is much, MUCH, better than 87%. I'll let you run the numbers, and then repost here what you find. If you'd like to discuss this further, I'll email you.
ReplyDeleteHi,
ReplyDeleteWhich broker do you use? Mind (IB) often does not have these share available for shorting.
Thanks!
@Matt Callow
ReplyDeleteInteresting suggestion! I'll try it. There are many optimizations that can be done to this strategy.
@T.
ReplyDeleteI'm using IB. You're right, I noticed shares aren't always available but, most of the time, if you can wait a few minutes, there are enough.
Interested to see new results and Variations. What software are you guys using to test your ideas?
ReplyDelete@Rok
ReplyDeleteI'm downloading the end of day data from yahoo and use Excel to run the numbers and draw charts.
You might be missing 1 thing. Proshares made a huge dividend on Dec 23. Look at a daily chart of TWM to see. It was nearly 35%.
ReplyDeleteThis came as a huge surprise to everyone holding their double-short funds. Being short TWM, you would need to cough up the cash to pay this dividend.
I don't believe FAS and FAZ have been around long enough to force a payout. But one morning they suddenly will.
* as a side note, Proshares screwed up the timing of the payment. They dropped the value of TWM on December 23, but didn't pay out the cash to holders for 7 days later. Massive margin calls ensued. Hopefully they will do a better job next time.
@AJG
ReplyDeleteThat's a good point. I had a look on earnings.com and it looks like there has been two dividends on FAS: $0.016 on 24-Mar-09 and $0.047 on 17-Dec-08. In both cases, they declared the dividend 1 day before the ex-date. Nothing for FAZ. If the dividend isn't too big, it doesn't look like an issue even if you're short; even if you have to pay the dividend, the stock will go down by the same amount.
As a follow-up, here is the info on all ProShares distribution payouts for 2008. The "Short Term Cap Gains" were the ones they paid in cash.
ReplyDeletehttp://www.proshares.com/funds/distributions/Distributions2008.html
I cannot find the Distribution info on Direxion's site. But it's something you'll need to look into.
@AJG
ReplyDeleteI found the distribution document from Direxion: http://www.direxionshares.com/pdfs/0309DisSchedule.pdf
Kaven - good research. Pennies in dividends are no big deal, you're right. I still can't get a straight answer why they had to pay out $20-$50/share on funds that were less than $100/share. That took everyone by surprise.
ReplyDeleteBut yes, fading these leveraged funds from time-to-time is a good strategy in itself.
@Matt Callow
ReplyDeleteAfter looking at your strategy, it looks very promising! But, on a long trending run, you will get a considerable draw-down. If you can weather those, you can make a huge profit but if you want more stability, you might want to rebalance anyway after a set percentage of deviation. Nice work!
in the long run this strategy will be profitable but when I say long run I mean loooong run. There's no free lunch here. You will have to be able to withstand huge drawdowns and commit further capital when one side of the market is trending otherwise you will risk geting margin called out at an adverse time.
ReplyDeleteWhen FAZ and FAS both debuted FAZ went from 60 to 200 while FAS went from 60 to about 11. Shorting both when they debuted would have resulted in a net loss of 150% in the first 3 weeks! A lot of people would have got margin called out of this position before it could have resulted in gains.
In addition, when one side of the market is steadily trending the results can be catastrophic because you can see one of these etfs gain about 1000% while the other tanks 98% resulting in a net loss of -900%!! This is what happened to the prices of the betapro double short and double long crude ets respectivly after the peak of the oil bubble last yr. This is an extreme example but as we have seen, markets do go to extremes from time to time.
ok, I just realized that the strategy involves rebalancing the positions to make the shorted ammounts equal each day. That definatley changes things. It would still require you to commit a lot of extra captial to withstand those times when one side of the market trends...and I haven't crunched the numbers to see if this actually would be as profitable as you say.
ReplyDeleteGood post but you risk getting your shares called in by your broker unexpectedly. Also, since this is by now a popular strategy, you risk not finding enough/any shares to short on a given day when you need to rebalance. Better to buy way in the money puts on both of them. Your risk is defined.
ReplyDelete@nodice
ReplyDeleteYou are right that this is not a risk-less strategy. A quick backtest shows that the biggest draw-down is around 5%. Of course, this isn't an absolute number; the draw-down may be bigger in the future.
@ghamal
ReplyDeleteYou're pointing something very true: shorting FAS and FAZ isn't always possible if you play with big amounts. Do you know if there is a delay between the short cover notice from the broker and the moment they actually force it? I assume they give you some time to close your position. Playing the deep in-the-money puts is worth looking into! Thanks for the tip.
I tried out a couple other pairs like SRS/SDS, ERY/EXY, BGZ/BGU, and DIG/DUG but none of them come any where close using this strategy as FAS/FAZ. I wonder why that is? From my backtesting it looks like there was a little one-time profit glitch at the beginning but if you start the backtesting a few days later, it give a profit of 3-4% a month with minimal drawdown. Slippage might be a problem but it might not be too bad since you only have to cover or short a few percent of one position each day. Not being able to get shares to short is probably the biggest concern.
ReplyDelete-=David
@sitemoney
ReplyDeleteYou're right, I observed the same thing on my end. It looks like FAS/FAZ is the best pair because it is the most volatile.
Be aware this strategy, if you simulate QLD and QID back to 2006, you will find that actually it will go up all the way to +10% until mid 2008, then go down. Since FAS/FAZ has little history, I am not sure what it would. Looks to me now that in the bull market, we get the opposite direction. I already have $20K in this strategy but I decide to take out next trading day unless I know what is the true reason. I still do not understand why this happens since I checked the ETF website and the leverage is from swap. If that is the case, there is no reason to have time decay. So the only possible explanation to me now is related to bull market or bear market. But I cannot prove it.
ReplyDeleteI started doing a variant of this strategy today, but came across something odd.
ReplyDeleteFidelity charges a 1% surcharge for shorting FAS, and a 3% for shorting FAZ.
Has anyone else experienced this?
What brokerages do you use to short these?
@Tony Zhou
ReplyDeleteThis is a good observation. I will look into it too. Also, for the past few days, it seems like there isn't a lot of decay. I think we have to look at the volatility index to know more.
@Paul
ReplyDeleteI use Interactive Brokers. They say they take a small percentage off the interest earning from the cash proceeds of the short. Since the US rate is near 0, I have paid some very small amounts of interests on my shorts, but it's near 0%. Maybe Fidelity charges more because of smaller short availability.