Ultra ETF's - Not a good investment?

or Register to post new content in the forum

10 RepliesJump to last post



  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Jan 1, 2009 5:25 pm

Just for trading and not for investing?


More ProShares Ultrashorts Tomfoolery
<DIV id=article_info extended="true"><SPAN id=page extended="true"><SPAN =authorname_for_print extended="true">by: Trader Mark December 29, 2008 | about stocks: <SPAN id=aboutstocks extended="true">EEM / EEV / FXI / FXP / PTR    

<P extended="true">The more I look at these double inverse ETFs the more I see how useless they are as even a medium-term hedging vehicle. It really leaves those of us who cannot short in an account (which is how Marketocracy.com is set up as are 401ks/rollover IRAs) in a pickle.<BR extended="true">
Some of them are hard to analyze directly since there is no specific index they directly short against, but the two foreign ETFs we own show how ineffective they are in any lengthier time frame. These truly are like options in which the time decay is working against you every day you hold them. I've already outlined how poorly they have done as hedges against financials or real estate dropping... here are some more cases.

I bought Ultrashort Emerging Markets (EEV) as a hedge against foreign markets dropping - I began this ETF on November 15, 2007 right near when it first began trading. I've generally held this in a 1-3% stake for most of the past 13 months, although at peak its hit 4-5% of the portfolio. At the time I originally bought, the instrument it hedges against - which is the<SPAN extended="true"> iShares MSCI Emerging Markets Index (EEM) was trading around $49. Today it is $24 - thats a 51% drop. I made an excellent theoretical call here, and the gain one could gain by <STRONG extended="true">shorting the long ETF - EEM would be +51%.

Below is what EEM looks like graphically since I bought its Double Inverse

<BR extended="true">In theory an Ultrashort is supposed to give you double that return - at least conceptually. In the prospectus and as documented by other what it guarantees is only each day's inverse return - i.e. the compounding is useless. But at least in concept if you held this ETF you'd think you'd get double the inverse return. Since the EEM fell 51%, you'd think you would have made +102%. Great trade! Not so much.

Ok we say - these instruements are flawed - instead of getting +102%, well maybe you only got 51% - which is not double the inverse but single the inverse the ETF is betting against. Or maybe it's even worse than that, maybe you only got half the inverse - maybe 25.5%. That would be awful but at least you still made money.

Want to see what the Ultrashort Emerging Markets (EEV) has done since bought on November 15th @ $60?<BR extended="true">
It is now trading at $56. Now granted there were some capital gains distributions, but at $56, it's a <STRONG extended="true">7% loss for holding it the entire time (ex capital gains) versus what should be (conceptually) a 102% gain. Let's say the capital gains distributions even led to a 5% gain. That's still rotten versus the concept of this ETF. So effectively, unless you catch the huge waves when this name really moves (or in fact any Ultrashort ETF) every day you hold this you lose money in a sideways market. You didn't get +102%. You didn't get half that - 51%. You didn't get a quarter of that - 25.5%. If you were lucky you got 0%. Despite making a great call on shorting foreign markets. That's terrible - I'm sorry.


I won't go through all the conceptuals with the Chinese ETFs but it's the same idea with iShares Xinhua China 25 (FXI) and its double inverse <SPAN extended="true">ProShares Ultrashort Xinhau China 25 (FXP). The latter is supposed to give you <STRONG extended="true">double the return of the index (but it only does on a daily basis). We bought FXP a few days after EEV - on November 20, 2007... or roughly at $58. If you were not around at the time October 2007 was the height of Chinese stock insanity[Aug 28: China "A" Shares Bubble] [Sep 1: The Growing Bubble in the Shanghai Index] - which we outlined daily - PetroChina (PTR) hit a $1 Trillion market cap [Nov 1 2007: PetroChina the 1 Trillion Dollar Company? Is *this the top?] , a multitude of Chinese small caps were going up 50% a day as speculators ran in and out of them, and many Chinese large caps had doubles or tripled in the year past.
<P extended="true">It was a bubble in the making that we were pointing out. The danger is you don't want to get in front of a bubble, but by November 2007 the "decoupling" theory (foreign markets will thrive even if the US goes to recession) was starting to show flaws and with this shiny new ETF, we had an easy way to bet against China. To nail a long term bubble within a month or two of its top is a heck of an achievement. Let's see how we were "rewarded"

Here is how the iShares Xinhau China 25 (FXI) chart looks from when we began buying the Double Inverse ETF - again it was roughly $58 when we bought. Today? Despite a huge rally in Chinese stocks the past month, it has only bounced to $28. That's a<SPAN extended="true"> 52% drop, so it looks like an excellent call!

But only if you shorted the FXI....<BR extended="true">
Because as you see below, our lovely ETF which should be double the inverse - again, should of (conceptually) netted us +104%, instead has fallen from $70 (where we started it in late November) to $40. (note: some capital gains are in that number so maybe the real price is $45, or $50). Let's be generous and say the capital gains added a whopping $10 (or 25% of the current value of the ETF).

So the double inverse has dropped from $70 to $50, or 28.5%, while the index it is shorting against has fallen 52%. So not only did we not get the double inverse (compounded) - 104%... not only did we not get a single inverse - 52%.... not only did we get HALF a single inverse - 26%... we did not even get 0%. We lost money. At $50, a "buy and hold" type would have lost nearly 30%...

Unless you are an extremely adept and nimble short term trader who held this witches brew on the exact right days/weeks and completely sold out all the other days, this is a loser. As is this whole concept of hedging by holding these vehicles - unless you are hedging for only a day or 3-4 days. And the market is going in exactly the right direction and in a very meaningful way.

Now that we have a historical record, we have to reconsider how to hedge; in a real mutual fund - I'd be happy to short the underlying indexes and it would make very large amounts of money for shareholders. Both these calls should have made investors 50%+. Frankly, even more than that since we've been trading them and had much higher stakes when they really fell off a cliff. Instead, despite some frantic trading to avoid the moves that can sap a month's worth of gains in 6 hours with these ETFs, I'm up only $10K on EEV and actually DOWN $4K on FXP. Despite both indexes I've been betting against falling 50%+ since I began these positions. I should be up $40-$60K+ in both considering the position size I've had these at, and how much they've fallen.

As a medium to long term hedge these are useless. They feed into the casino mentality and can make big bucks in very short periods of time which makes the gambling types in the market happy (along with run and gun hedgies), but for the greater investment community who are looking for a way to buy insurance for the long term against your long positions - the history shows they are inappropriate. As more suckers... err people, like me figure this out, I believe these tools will become abandoned except for the extremely short-term oriented crowd and/or during times of extreme duress in the market.


On a personal note, I was aware of the issue here, but vastly underestimated how much this issue effected the long-term performance. I did realize they only gave double the inverse on a daily basis, but I though I'd still catch anywhere from 60-70% of the move once compounded (and I'd give up 30-40% of the gain) due to this liability. I was willing to give up 1/3rd of my gains since I have no other way to short - as long as I was correct directionally I'd still make money; that was my thinking. Instead I see I am giving up ALL of the move (and in fact one is prone to lose money even if "correct") if your holding period is of any length of time. This was the flabbergasting point. Due to this, I'll be cutting back these positions severely because in a sideways market, they steal from you each day; I'll only use them in strongly downtrending periods. For a long term hedge against long positions, I am not sure there is anything out there of use for those who are stuck in "long only" accounts.

p.s. I've taken a quick look at the Ultra Long versions of these ETFs and they have the same problems... I would make an argument that if you really want to bet against a sector, even better than shorting the underlying Index is short the Ultra long ETF. Not only do you get the drop from the underlying index, but you get the Proshares ETF "Ultra" or "Ultrashort" structural degradation I've outlined above.
<P extended="true">So if I could, and I wanted to short Real Estate - short Ultra Real Estate (URE) which would give you a much better return than shorting the underlying index iShares Dow Jones US Real Estate (IYR). Want to see it in numbers? Since November 2007 IYR has fallen from $75 to $35 (a drop of 53%). URE? From $47 to $6 (a drop of 87%). So you get to benefit from the underlying weakness of the sector plus the flaws of the ETF for anyone who holds it medium term or longer. The pain you might suffer when the underlying stocks rallies is compensated by the breakdown in the structure of the ETF. I am beginning to wonder if due to the structure if all these ETFs are destined for a near $0 price in the "long term".

Jan 1, 2009 7:11 pm
Understanding ProShares' Long-Term Performance

Understanding Long-Term Performance | Performance | FAQs on Performance and Pricing

ProShares are designed to provide either 200%, -200% or -100% of index performance on a daily basis (before fees and expenses).

A common misconception is that ProShares should also provide 200%, -200% or -100% of index performance over longer periods, such as a week, month or year. However, ProShares' returns may be greater than—or less than—what you’d expect over longer periods.

How does this happen?

The hypothetical example below demonstrates how ProShares' long-term returns may be different than what you'd expect:

Fund XYZ seeks to double the daily performance of Index XYZ. On each day, Fund XYZ performs in line with its objective (200% of the index’s daily performance). You might expect that over the entire five-day period, the fund’s total return would be double that of the index, but that’s not the case. For the entire period, Index XYZ gained 5.1% while Fund XYZ gained only 9.8%.

Why does this happen?

This is due to several factors, but a significant one is index volatility and its effect on fund compounding. In general, periods of high index volatility will cause the effect of compounding to be more pronounced, while lower index volatility will produce a more muted effect.

To demonstrate this point, let’s compare Fund XYZ from the previous example to the less volatile (and also hypothetical) Fund ABC, designed to double (200%) the daily performance of Index ABC.

At left, the steady uptrend of Index ABC led to a compounding effect (magnified gains upon gains) in Fund ABC that more than doubled the index’s return (10.4% vs. 5.1%). At right, the more volatile Index XYZ led to magnified fund gains and losses, producing less than 200% (9.8% vs 5.1%).

(Remember, you can’t invest directly in an index, and hypothetical fund performance does not reflect fund fees and expenses.)

What it means to you

In the end, ProShares are designed to accomplish their objectives on a daily basis. As a result, you shouldn't expect ProShares to provide 200%, -200% or -100% of index performance over longer periods.

If you're interested in more details about ProShares' long-term performance, please see the description of correlation risk in the ProShares prospectus. Also, you can download the ProShares statement of additional information for a deeper discussion of the factors that affect ProShares' long-term performance, including a tool that estimates one-year fund performance based on index behavior.

A note about daily benchmarks

A fund could be designed with a benchmark that uses a time period longer than one day (for instance, a week, a month or a year), but index volatility and its effect on compounding will create a similar effect over multiple periods, regardless of their duration.

Jan 1, 2009 10:19 pm

That was way too much to read...

Jan 2, 2009 10:41 am

Yes, but if you did, maybe you learned something. 

Or, like me, maybe not.
Jan 2, 2009 10:52 am

Oh I read it, but don't like the idea of the Ultras...

Jan 6, 2009 4:31 pm

Interesting.  I'm not into ultras too because I think that I won't know how much and what I'm shorting or doubling.  I may just end up ultra-shorting an ultra-long.

Jan 6, 2009 5:01 pm

Also, why would anyone buy an ETN?  I cannot understand why. The people who bought the Bear Stearns ETNs are also wondering why they bought them.

Jan 23, 2009 1:38 pm

I road the DXD from (december 07) 13300 to 8000..since I have been buying around 9K and selling at 8K..Right now i am long 2x inverse at 8100

Jan 23, 2009 11:37 pm

Read the prospectus, it clearly states the goal is only daily. 

Jan 24, 2009 5:50 pm

Here again if you don't know as assets delta you c an't effectively hedge it, and only knowing delta gets you a little way there, because you've also got gamma (delta's second derivative), and vega (volatility), tau (time), rho (now there's two kinds of interest rate risk - yield cureve shifts and yield curve rotataions). Unfortunately you can calculate delta yourself but its probably better to have some options trading software do it. You're dealing with the calculas here so the actual computed numbers are only acurate for tiny changes in underlying prices / rates / spread.

ProShatres should be faulted for calling this stuff / selling it as double hedges, because they only work in the event volatility is zero.