Opinion

Felix Salmon

ETFs aren’t derivatives

By Felix Salmon
October 1, 2010

Herb Greenberg misreads the SEC flash-crash report, in an attempt to justify his silly claim that ETFs are derivatives. I agree with Herb that there are serious questions which should be asked about ETFs, especially ones which include small-cap stocks. You can’t turn an illiquid stock into a liquid stock just by throwing it into an ETF, and illiquidity is one of those things which goes hand-in-hand with volatility and attempts at price manipulation.

But that doesn’t mean that ETFs are derivatives. They’re not.

Herb quotes this bit of the SEC report:

The E-Mini and SPY are the two most active stock index instruments traded in the electronic futures and equity markets. Both are derivative products designed to track stocks in the S&P 500 Index, which in turn represents approximately 75% of the market capitalization of U.S.-listed equities.

There is a way in which this is true, but it’s best read as a slightly clumsy attempt to lump the E-Mini and the SPY together with some inartful phrasing.

The E-Mini is a derivative product by dint of being a derivative. It’s a futures contract, a zero-sum game, an instrument whose value at expiry is a wholly transparent function of the value of some other financial instrument.

The SPY, by contrast, is a derivative product only by dint of the fact that it’s a product — a security, not a derivative — which is derived from aggregating 500 different stocks. You couldn’t have the SPY without the S&P 500, so in that sense the SPY is derived from the S&P 500. But if you own shares of SPY, you have real wealth: real claims on real assets of 500 real companies in the real world.

Think about it this way. If all of the shares of SPY were to become vaporized tomorrow, then the S&P 500 itself would rise, since the total number of shares outstanding in the stock market would have fallen. Since the value of those 500 companies wouldn’t have changed, the value per share would be higher.

If all of the world’s E-Mini contracts were to become vaporized tomorrow, by contrast, then the effect on the S&P 500 would be de minimis. E-Mini contracts are side bets on the S&P 500: they’re not real-world claims on it.

What the E-Mini and the SPY have in common is that they’re both trading vehicles: both of them are used to make bets on the short-term direction of the S&P 500. That’s why, pace Greenberg, there’s so much short interest in SPY. If I want to short the S&P 500, I’ll borrow a few shares of SPY and then sell them in the market. Those shares will be bought by my counterparty, who will then turn around and lend them to someone else who wants to short the S&P. And so on and so forth: the same shares can end up being shorted many times by many different people.

But that doesn’t make them derivatives.

The distinction between securities and derivatives is a useful one. It’s always possible to argue that the value of any financial instrument is ultimately derived from the value of some other financial instrument: if you wanted to, you could probably come up with a colorable argument that senior unsecured bonds are derivatives of stock options. But in the real world, stocks aren’t derivatives, and ETFs are stocks. Derivatives, meanwhile, are something you call financial instruments you don’t like. Which is why occasionally people will erroneously say that CDOs, say, are derivatives. They’re not, any more than ETFs are.

Comments
13 comments so far | RSS Comments RSS

You nailed it with this one line:

“But if you own shares of SPY, you have real wealth: real claims on real assets of 500 real companies in the real world”

Posted by KidDynamite | Report as abusive
 

I wouldn’t get so worked up about terminology, Felix. “Derivative” means different things to different people, and it’s hard to get an airtight definition. But one thing that makes absolutely no sense is your distinction between derivatives and securities; that is obvious from the common phrase “derivative security.”

Is a derivative necessarily derived from another asset? No. VIX futures are derivatives by any sensible definition but realized volatility is not a traded asset and there is no replicating portfolio.

Is a derivative necessarily a bilateral contract rather than a negotiable security? No. Exchange-traded options are negotiable but they are definitely derivatives.

Is a derivative necessarily something that has no real claims on real assets? Not even close. Plenty of derivatives are physically settled with the underlying asset.

Is a derivative necessarily a side bet? Of course not. It is often attractive to structure derivatives with zero initial cost, but this is hardly a precondition. Plenty of derivatives are funded, including the ABS/CDO tranches you like to write about.

Posted by Greycap | Report as abusive
 

A derivative is something whose VALUE is derived from the value of another asset.

A call option has value only becasue the price of the underlying asset is priced where traders are willing to pay something to buy the call.

SPY is not a derivative.

It’s essentially a mutual fund. SPY has value because it owns assets.

Mark
http://blog.mdwoptions.com

Posted by MarkWolfinger | Report as abusive
 

MarkWolfinger- I was with you when you said a derivative is something whose VALUE is derived from the value of another asset (such as a basket of S&P stocks)

I lost you when you said that an ETF whose VALUE is derived from the VALUE of the S&P basket was not a derivative though… because it’s “essentially a mutual fund”? That means it can’t be a derivative I guess… well okay then.

Posted by johnhhaskell | Report as abusive
 

Your claim may be specifically true with respect to SPY, but to make the blanket claim that ETFs are not derivatives is dangerously misleading. Somewhere around half of all AUM in ETFs in Europe is via synthetic replications – derivatives with no claim on the asset. And yes, they are sold as ETFs.

Posted by IbexSalad | Report as abusive
 

by that Greenberg’s logic even mutual funds are derivatives!

Posted by takloo | Report as abusive
 

Felix right, Greenberg wrong, end of story. Quite frankly I’m surprised it’s even an issue for debate.

Posted by ottorock | Report as abusive
 

I see the distinction Felix is making, but it seems a little nitty. Is there any difference between SPY leading to more speculation and S&P500 futures doing the same thing?

Posted by Mr.Do | Report as abusive
 

“If all of the world’s E-Mini contracts were to become vaporized tomorrow, by contrast, then the effect on the S&P 500 would be de minimis. E-Mini contracts are side bets on the S&P 500: they’re not real-world claims on it.”

Why do so many believe this? E-minis and SPY converge in price because they are near-perfect substitutes as investment vehicles. The only difference is that E-minis are discounted due to not receiving dividends. They’re very close substitutes, and economics tells us that prices and quantity of substitutes cause changes in the prices of each other.

If I want to bet the S&P 500 will go down, I have two options – either short SPY, or short E-minis. If E-minis go away, I will be more likely, and pay more in fees, to short SPY, causing the price of SPY to go down.

Likewise, if I want to bet that the S&P 500 go up, I have two options – buy E-minis, or buy SPY. If E-minis go away, I will be willing to pay higher fees to get SPY instead, and cause its price to go up.

In other words, if I own long E-minis, I have real wealth – I have claims on hedge funds who have bought SPY and shorted E-minis in order to bring their prices in line.

Posted by jmalicki | Report as abusive
 

This has been debated in the commodity realm for a long time, when Live Cattle became cash-settled. By your definition, Felix, cash-settled futures are derivative, because they do not represent a claim on a real asset. Physical settlement would not be a derivative, because you can alway take delivery of Pork Bellies.

But as several commenters above sense, this is a distinction without a difference. What matters is the nature of the underlying instrument. Otherwise you end up saying nonsensical things like, Bond Futures aren’t derivatives, because you can take delivery of the bonds, but Eurodollar futures are, because they are “side bets” (cash-settled)

If the underlying market is valid and liquid, the derivative should be meaningful, if volume in the two markets is balanced. But if there is a massive disparity in size or liquidity, as it is in the market for cash-based muni debt and the muni CDS market (to pick an egregious example), you end up with one or the other market being a video game for hedgies.

And if policy errors result from bad price signals, don’t say you weren’t warned.

Posted by Publius | Report as abusive
 

Actually many ETF’s may not hold the underlying equities, but provide equivelant returns versus a note or OTC contract. Most ETF’s refuse to disclose the counterparties of these notes. They are in effect daily swaps derived from the value of indexes or baskets. These are most common in 2 and 3X products which leads to their long term tracking errors as they are settled daily and suffer from the ETF decay issues induced by Siegels paradox.

One could argue the notional principal isn’t at risk as it isn’t swapped, but counterparty risk on the daily settlment is at risk. The notional values underlying the ETN instrument based derivatives may never actually trade, thus making the variance a derived value swap.

Posted by Nick_Gogerty | Report as abusive
 

The principal functional difference between the e-mini and SPY methods of speculating on that index is leverage. That is the important distinction to make.

Some ETFs are derivative-like, some are not. Some ETFs are structurally doomed (UNG comes to mind); some are comparable to mutual funds (SPY, e.g.). They are difficult to generalize, functionally.

Posted by mutant_dog | Report as abusive
 

So. ETFs for Dummies….which ones are considered scarey?? There must be some that are secured with real commodities, e.g. SGOL

Posted by deblythe | Report as abusive
 

Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/
  •