Arbitrage Opportunity in Ultra ETFs

Since 2006, ProShares has offered Ultra ETFs that have exposure to twice the daily exposure to certain underlying indexes. These ETFs can either be long or short the daily index. These ETFs are able to track the daily inverse by using financial instruments that allow them to use leverage.

ProShares Ultra S&P 500 (SSO) assets comprise of:

• $2.436 billion in S&P 500 Swaps
• $.392 billion in S&P 500 EMINI Futures
• $2.351 billion in the underlying equity of the S&P 500 index
• $.238 billion in cash

ProShares UltraShort S&P 500 (SDS) assets comprise of:

• -$6.568 billion in S&P 500 Swaps
• -$.519 billion in S&P 500 EMINI Futures
• $3.543 billion in cash

The problem with these Ultra ETFs is that their long-term performance will diverge from the underlying performance of the index that they track. The reason for this divergence is the fund’s use of leverage and volatility in the underlying index. The leverage used in the funds successfully allows them to track the daily price movements of the indexes that they are supposed to track, but leverage will also play against them over multiple periods if the index is extremely volatile. The charts below illustrate the problem with these ETFs. The SSO tracks twice the daily movement of the S&P 500.

As the above charts demonstrate, SSO will successfully achieve 2X the daily price movements and 2X long-term price movements only when the underlying index has successive returns. If the index demonstrates any volatility, the returns of SSO will diverge from 2X the index.

Both the SDS and SSO have a downward bias due to volatility in the markets. The below chart illustrates the potential divergence given large market volatility.

Empirically, these indexes have displayed these divergences over the past several years. An investor can take advantage of the effects that volatility has on these ETFs by constructing a portfolio that would leave an investor with theoretically no daily volatility. As displayed earlier, SSO is approximately half invested long in swaps and futures and half in the underlying equity of the S&P 500 index. SDS is only short swaps and futures in the S&P 500 index.

An example of the arbitrage portfolio would consist of shorting $100 SSO, shorting $50 SDS, and buying $50 of the S&P 500 index. This would leave an investor with approximately a $50 short exposure of long S&P 500 swaps and futures, a $50 short exposure of short S&P 500 swaps and futures, a $50 short exposure of the equity of the S&P 500 index, and a $50 long exposure of the S&P 500 index. The short-short and short-long exposure of the swaps cancel each other out to be market neutral, and the short-long and long exposure to the equity of the S&P 500 cancel each to be market neutral as well.

Constructing this portfolio would leave an investor with small daily price changes but gains in the ending of the portfolio. Looking at the hypothetical situation above, an investor could expect to make (.5*5% + -1* 8.13% + -.5*-14.69%) or 1.715% plus the interest earned on the short margin.

Because this strategy involves $150 in short and $50 in long exposure, an investor can earn the risk free rate on the excess cash. This strategy essentially allows an investor to borrow free money to invest elsewhere.

The line graph below demonstrates the returns that this arbitrage portfolio has displayed over the past several years (Line graph shows adjusted closing prices). During 2007 this portfolio (purple line) performed poorly, because there was little volatility and only upward movements in the S&P 500 index. Once the markets became volatile in 2008, the portfolio turned positive and has returned upwards of 40% over the past 3 years.

Given the recent uncertainty in the financial markets, this portfolio may be attractive for an investor seeking returns that are not correlated with the market.

Disclosure: Author currently has no exposure in this strategy, but will likely in the near future

Comments

[mule65:]

Very clever.

May 05 08:26 AM

[Augustus:]

I sure like the concept. Do you have any suggestions related to possibly rebalancing the portfolio at different stages of imbalance?

May 05 11:23 AM

[GoochyPuppy:]

Aren't the ETFs you mention almost always "difficult to short" (i.e., very few shares available at brokers to short). Shorting has become more difficult since nakeds are not permitted anymore. No?

May 05 12:02 PM

[snoopyjc:]

I was thinking of doing a similar thing, using FAZ and FAS. Initiating the short positions when either one has a "huge" move.

Have you considered doing this using triple-leveredged funds?
--joe

May 05 12:18 PM

[snoopyjc:]

How about shorting $150 BGU, shorting $100 BGZ, and buying $50 IWB?
--joe

May 05 05:21 PM

[Mad Hedge F...:]

This is a potential blow up. We are gathering a head of steam towards our next financial crisis, even before the current ones are solved. The perpetrator will be new financial product du jour, the super leveraged Exchange Traded Funds (ETF’s), which are being created at a breakneck rate, sucking in billions of dollars from investors. ProShares has filed for 94 funds, which offer traders 300% long or short plays in markets as diverse as the Russell 1000 Index, the MSCI Malaysia Index, and the Nikkei 225 stock average. Direxion has gathered $3.4 billion with 16 different 3X funds launched since November. There are now more than 800 ETF’s, and I have been a big fan of those for emerging markets (EEM) and short Treasuries (TBT), which allow investors to take positions in niche sectors and foreign markets which are otherwise difficult or expensive to get into. These also allow mutual funds the only means to go short, and include tax advantages and hedging opportunities. But the leveraged versions include risks that most buyers don’t fully understand, even if they parse through the voluminous prospecti with a magnifying class. They promise their triple tracking only for the day you buy it. Beyond that, the tracking error can be huge. The mechanics of these funds force them to be buyers of every rally and sellers of every dip. Over time, leveraged short funds can actually suffer large losses, even in falling markets, and vice versa. It is just a matter of time before one of these goes to zero, wiping out investors. They are already being blamed for an increase in market volatility in the last hour of trading. Gaming sector ETF’s has become the new blood sport for nimble hedge funds. You can expect a replay of a movie you’ve seen before. At the first sign of trouble, liquidity will disappear, auditors will mark them down to nothing, and suddenly the whole world will be for sale. Sound familiar? You have been warned!

May 05 05:46 PM

[Luck-o-the-...:]

Do the math...long $50 SPY is long $50, short $100 SSO is like being short $200 SPY, and being short $50 SDS is like being long $100 SPY, so you are NET short in this portfolio, which is the real reason for the decline in 2007 and the rise in 2008 ($50+$100-$200= - $50).
However, this idea holds merit. We use a similar setup like this in our hedge fund as part of the strategy; however, there are overlay and advantages when adding in option arbitrage trades (InterETF spreads) that we have also pioneered. Obtaining the shares to short is VERY difficult. Fortunately, we have been able to locate and obtain shares. If any qualified investor would like to know more, please shoot me an email.

May 05 05:54 PM

[Luck-o-the-...:]

Pick target rebalancing or use a stop loss.


On May 05 11:23 AM Augustus wrote:

> I sure like the concept. Do you have any suggestions related to possibly
> rebalancing the portfolio at different stages of imbalance?

May 05 05:57 PM

[Luck-o-the-...:]

This would create a short position as BGU equals -450, BGZ equals +300, and IWB equals +50.
You would create a net short position of $100 on the IWB, so you better be bearish.


On May 05 05:21 PM snoopyjc wrote:

> How about shorting $150 BGU, shorting $100 BGZ, and buying $50 IWB?
>
> --joe

May 05 05:58 PM

[Luck-o-the-...:]

One other caveat...your cash is not "free" These are hard to borrow securities, and you WILL pay interest on your short positions. In some cases, it will be large as compared to any short interest rebates you may be able to obtain. The margin maintenance on some of the leveraged short positions, especially 3x, will be 100% of the short value, so you are not able to use that cash for much, if anything. We have been doing this along with options strategies around it for quite some time, and know the actual costs. Also, you need to be very patient, and on top of it. The strategy does not look very good when the market goes up for 8 straight weeks either. If you aren't familiar with it, you are bound to capitulate and rebalance at the most inopportune time.

May 05 06:01 PM

[Luck-o-the-...:]

Robert,
Shoot me an email (tcollins@claruspartne... I'd like to talk with you further if you have any interest.

-tim

May 05 06:02 PM

[PutneyHillYank:]

Give it a rest if you have assets in equity markets you like but need some comfort then buy some BGZ its that simple.

May 05 07:08 PM

[PutneyHillYank:]

Great I am just saying that the day Congress dropped the ball and we dropped 700 I wish I had BGZ on.


On May 05 06:02 PM Luck-o-the-Irish wrote:

> Robert,
> Shoot me an email (tcollins@claruspartne... I'd like to talk with
> you further if you have any interest.
>
> -tim

May 05 07:20 PM

[PutneyHillYank:]

I will say that the 2 cent miss by CHK resulted in my opening a long postion in that one when it was down 12 percent for the day this afternoon. But trading is easy just fade the crowd when they are in extreme mode.

May 05 07:24 PM

[Luck-o-the-...:]

Not even related to the topic. What gives??


On May 05 07:24 PM PutneyHillYank wrote:

> I will say that the 2 cent miss by CHK resulted in my opening a long
> postion in that one when it was down 12 percent for the day this
> afternoon. But trading is easy just fade the crowd when they are
> in extreme mode.

May 05 08:02 PM

[Luck-o-the-...:]

What is your point?


On May 05 07:08 PM PutneyHillYank wrote:

> Give it a rest if you have assets in equity markets you like but
> need some comfort then buy some BGZ its that simple.

May 05 08:03 PM

[squark62:]

interesting post. i arbitrage SDS-SSO, DXD-DDM & QID-QLD in pairs by re-balancing the ETFs themselves, no options or shorting of shares. my model holds the ETFs indefinitely as long as measured boundary conditions are favorable. the pairs are allowed to float until a re-balance signal indicates re-weighting the pair to delta neutral or ultra bias the pair for high beta. the model computes daily EOD neutral weights and another indicator tells which ETF ultra bias. backtesting the most updated revision of the ultra bias scenario indicates an SDS-SSO pair would be up 88% YTD 2009.

May 05 10:27 PM

[Robert Zingale:]

Luck-O-Irish I see what you are saying about the exposure. I was originally just looking at the asset allocations within the ETF to construct the portfolio and forgot how the returns on SSO are actually not completely covered under my article. It would be better to replace the $100 with only $75. This would still leave the strategy at a 28.15% return when I wrote the article.

As for what you are saying about the cash not being free, you may be completely right. I have not attempted to complete this transaction yet.

May 06 01:00 PM

[DanielC:]

I've been trying to do this kind of thing for a rather long time. Very hard to find SSO and SDS to short, though (at least for the retail investor). I do have some short positions on these, and I've never been called to cover. However, many times I tried to add to my short positions, and I got the "we could not find the security" message. It gets frustrating after awhile.

May 06 03:31 PM

[Luck-o-the-...:]

We do this, and have for years, and I can tell you with certainty on the cash that what I said is correct at the current moment. It doesn't make it a bad strategy though, and I never intended the comment to imply this.


On May 06 01:00 PM Robert Zingale wrote:

> Luck-O-Irish I see what you are saying about the exposure. I was
> originally just looking at the asset allocations within the ETF to
> construct the portfolio and forgot how the returns on SSO are actually
> not completely covered under my article. It would be better to replace
> the $100 with only $75. This would still leave the strategy at a
> 28.15% return when I wrote the article.
>
> As for what you are saying about the cash not being free, you may
> be completely right. I have not attempted to complete this transaction
> yet.

May 06 05:00 PM

[Zac:]

Irish,
But in the above example the S&P went up 5%, so if he was net short wouldn't he have lost money as opposed to returning 1.715%?
Thank you for your posts, very interesting stuff.

On May 05 05:54 PM Luck-o-the-Irish wrote:

> Do the math...long $50 SPY is long $50, short $100 SSO is like being
> short $200 SPY, and being short $50 SDS is like being long $100 SPY,
> so you are NET short in this portfolio, which is the real reason
> for the decline in 2007 and the rise in 2008 ($50+$100-$200= - $50).
>
> However, this idea holds merit. We use a similar setup like this
> in our hedge fund as part of the strategy; however, there are overlay
> and advantages when adding in option arbitrage trades (InterETF spreads)
> that we have also pioneered. Obtaining the shares to short is VERY
> difficult. Fortunately, we have been able to locate and obtain shares.
> If any qualified investor would like to know more, please shoot me
> an email.

May 06 05:55 PM

[Luck-o-the-...:]

It is not the end result you care about, it is how it gets there. Just because you start net long, net short, or neutral, if enough time passes, you could make money in any of the starting positions based on how the market got there. However, if you look at Mar 9th closing to now, and you take this scenario of going long $500 SPY, short $1000 SSO, short $500 SDS for a net short credit of $500, you would have bought 7.4 shares of SPY, shorted 69.1 shares of SSO, and shorted 4.34 shares of SDS. Since we are in a trending market, you would now find yourself sitting with a liability of roughly $1378 or -1378. Remember you only got $500 on the sale, so you are down 878. For some people, this literally could have been their ENTIRE account. Since you can margin these, and you have short positions, and believe me you ARE paying interest, you are not getting any, there are risks here. In a trending market, the compounding can hurt you as much as it can benefit you. You have to play this kind of game with a deep wallet or using no margin.


On May 06 05:55 PM Zac wrote:

> Irish,
> But in the above example the S&P went up 5%, so if he was net
> short wouldn't he have lost money as opposed to returning 1.715%?
>
> Thank you for your posts, very interesting stuff.
>
> On May 05 05:54 PM Luck-o-the-Irish wrote:

May 06 06:12 PM