Felix Salmon

Securities are not derivatives

By Felix Salmon
June 1, 2009

The WSJ makes a good catch: the SEC — the agency charged with regulating securities but not derivatives — refers to securities as derivatives three times in a press release which came out last Thursday. This is typical:

Rosalind R. Tyson, Director of the SEC’s Los Angeles Regional Office, added, “These brokers took customers primarily interested in protecting their money and pushed them into risky derivative investments through blatant misrepresentations.”

No, they were not derivatives, they were mortgage-backed securities, as the official complaint — which never uses the word “derivatives” explains.

When I’ve made this point in the past, invariably people have popped up in the comments saying that these securities can indeed be considered derivatives, for some recondite etymological reason. But really the line between securities and derivatives is quite easy to understand: derivatives are a zero-sum game, while if a security goes to zero, no one else makes a concomitant profit.

I do wonder, though, whether a lot of the animus aimed at derivatives comes from people who are using such a broad definition of the term that it encompasses mortgage-backed securities and other non-derivative instruments. Certainly it would be helpful if stories like this one were assiduous in defining their terms — I’m pretty sure it’s referring only to real derivatives and not to securitizations, but that’s a distinction which is evidently worth making explicitly. And it’s certainly a distinction which the SEC, of all agencies, should be hyper-aware of.

19 comments so far | RSS Comments RSS


From my quick read of the complaint, it looks like the instruments were CMO’s derived from RMBS. That makes them derivatives. Typically, RMBS were registered for a year, using disclosure based on unstructured non-standard ASCII text, and then deregistered; derivatives created on top of those “temporarily public” or “psuedo-public” securities were typically exempt from registration because they were only supposed to be purchased by “sophisticated” investors. (Sophisticated! Ha! Participants in the shadow banking system may have been many things, but apparently only the folks who were short the RMBS, CDOs, CMOs, etc. were truly sophisticated. Good Wired story linked from http://paulwilkinson.com explains that sophistication required a third party to manually structure the data, translating it from ASCII to XBRL, to detect the defects in such investments.)

The SEC’s enforcement authority isn’t affected by semantics. Generally, it extends to any “security,” defined as an investment of money in a common enterprise for profit derived from the efforts of others. There are specific exemptions, but promoters can’t hide behind mere jargon such as “derivatives.” There’s much more at http://paulwilkinson.com if you’re interested.



Um, someone has made a ‘concomitant’ gain…the guy who took out the mortgage in the first place and now hasn’t paid it back…

Posted by Pedant | Report as abusive

The distinction is not very clear cut. E.g If a company issues warrants on its own stock this is generally thought of as a derivative (as it clearly derives its price from the underlying equity) despite being a security by your definition.

Posted by Tim Bassett | Report as abusive

I work on a securitized derivatives desk. Everything we trade is both a security and a derivative.

Posted by Sean | Report as abusive

Please open your basic finance textbook, or a basic derivatives textbook. Now turn to the glossary and look up the word “derivative.” Do you find your definition?

Before you close that book, look up the word “security”. Pretty broad isn’t it?

Using those standard definitions, all derivatives are securities. Also, all securitized products, such as CDOs, are derivatives by those definitions.

Not surprisingly, if you make up your own definitions, you’ll find that not everyone else uses words the way that you think they should.

Posted by Simon | Report as abusive

Sorry, but a mortgage-backed security is a derivative. A derivative is defined as an asset whose value depends upon the value of another asset. In this case, the MBS value depends upon the value of the underlying mortgages.

Posted by clayton | Report as abusive

Clayton is correct.

A derivative is an instrument that is derived from one or more other instruments.

http://en.wikipedia.org/wiki/Derivative_  (finance)

Where do you get your definition from?

Posted by dk | Report as abusive

Can you explain how a derivative is zero sum but a bond isn’t?

Can you explain how counter party risk figures into zero sum derivatives like CDSs?

Can you explain how a bundle of synthetic MBS (made up of zero sum derivatives) is not a derivative?

These aren’t rhetorical, I feel I missing something.

Posted by zach | Report as abusive

The definition I used is from Hull’s “Options, Futures and Other Derivatives.”

Posted by clayton | Report as abusive

Ah – another fun flame-fest about terminology. The problem is that there are no internationally-recognised bodies issuing universally-accepted definitions of technical jargon for the world of finance.

However there are some commonalities in the definitions. I have never before seen anyone give a definition of security that didn’t require it to be a negotiable financial instrument. Therefore a bilateral bank loan contract would not be a security, but an otherwise identical bond would be. If the bank securitised its exposure via a credit-linked note, the CLN produced would be a security. That’s kind of what the word securitisation was invented for. A derivative may or may not be a security depending on its form (eg futures are, bilateral Contracts-For-Difference aren’t).

Defining derivative is even more fun. I don’t think the definition Felix gives is adequate. After all, what is the difference in cash-flows between a collateralised CFD and a highly leveraged direct investment in an asset? Both require an up-front contribution of cash, regular interest (or “notional interest”) payments and then a settlement cash-flow at termination. Almost everyone would call a CFD a derivative, and few people would call the leveraged investment a derivative. This is not captured by Felix’s definition.

I know little about US regulatory law, but isn’t there some sort of difficulty for the SEC in regulating pure derivatives, as opposed to securities (the word that is in their name)? Possibly the items in question could be described as both. For the legal proceedings they described them as securities because that is in their enforcement scope under relevant law, but for the press release they chose the more commonplace description of calling them a derivative.

Posted by JH | Report as abusive

A Derivative is an item whose value is determined by reference to an (ultimately, in some cases) underlying security.

For purposes of regulation under the 1933 Act, it is therefore both a derivative and a Security.

Posted by Dollared | Report as abusive

Read the SEC complaint. These were derivatives in the most common sense of the word. They were collateralized by Mortgage Backed Securities. It is the press release’s headline that is misleading, and it is the WSJ writer who is confused.

An example from “The Types ofCMOs Traded in the CMO Program”:

“Inverse Floaters are variable rate securities with a coupon that is inversely related to a short-term interest rate index, typically the London Interbank Offered Rate (“USOR”). As the index’s interest rates rise, the Inverse Floater’s interest payment falls, and vice versa. Inverse Floaters can have poor liquidity and erratic pricing. Inverse Floaters purchased for a premium (i.e., at a price over par) or sold before maturity present price risk to investors (i.e., the investor can lose their original investment).”

Posted by S Bayer | Report as abusive

I agree with Felix that a garden-variety MBS is not a derivative as such – unless you expand your definition of “derivative” to include all structured finance as well, but then the term is too vague to be useful. I think Felix’s definition of “derivative” is pretty good actually. Unfortunately, the press seems to have adopted the term “derivative” as shorthand for any sort of guilty-until-proven-innocent Wall Street hocus-pocus.

Posted by Nate | Report as abusive

These words have one definition for legal purposes and another definition in finance textbooks. I suspect that this is true of many other financial products too.

Posted by Anonymous | Report as abusive

I agree with Nate.

You can follow this down the rabbit hole. A stock is simply a call option on firm value with the debt as the strike price. A mortgage is a risk free rate with an option to default (call option on household) and an option to prepay (put option on household). Are mortgages and stocks derivatives?

Zach – a good key is to follow ownership. I buy a GE bond, I loan money to GE. I buy a GE stock, I own a part of GE. I buy a CDS and a call option on GE’s bond/stock, I’ve made a bet with a third party related to how GE will do. I’m not transacting with GE.


S Bayer is right here. I’ve actually traded this product and these bonds are referred to as “mortgage derivatives.” That’s because the coupon is set based on where libor goes, hence the value of the bond is a derivative of libor. It also includes interest and principal only strips, which aren’t obviously derivatives, but are far enough away from the normal cashflows of CMOs that they fall into the mortgage derivative category as well.

I agree that the definitions of derivatives and securities are not super clear, but there are widely accepted ways to refer to these securities. There are mortgage pools, which is a collection of mortgages with similar characteristics. These pools go into a CMO, which will have tranched up cashflows. Some of these bonds will have a fixed coupon, but have special pay rules about the order they receive principal (first, second, according to some schedule, etc). Those are referred to as fixed rate CMOs. Other bonds, such as the ones in the SEC complaint, have coupons which move with respect to libor and those are called mortgage derivatives. If you call the fixed rate CMOs derivatives you’ll just confuse everyone. The value of a stock is dependent on the performance of the underlying company, but that doesn’t mean it should be classified as a derivative (although it can sometimes help to think of it as a call option on a company).

Here’s the part that I don’t get in the complaint: they were levering these guys up 10:1! That’s insane. These are bonds with base case yields of 15%-25% and option adjusted yields between 8% and 20%. You’re getting stock market level returns without leverage. Probably better. If they hadn’t levered up these positions most of these people would be doing much better than if they had invested in the stock market. In fact, they probably would have had phenomenal returns if they hadn’t been forced to liquidate because of margin calls. Either way though, these are very risky, complicated bonds and it sounds like they tried to pretend like they were conservative products. There’s a reason you make 20% a year – it’s because you can lose a lot of money very, very quickly.

Posted by a reader | Report as abusive

For proof positive about how ridiculously confusing the terminology is, check out Ezra who cites you approvingly: http://voices.washingtonpost.com/ezra-kl ein/2009/06/derivatives_defined_thrice.h tml

and then goes on to call CDOs derivatives: “For a longer — and very useful — introduction to credit derivatives (like credit default swaps and collateralized debt obligations, both of which had a particularly central role in this crisis) …”

Posted by Anon | Report as abusive

You should split the world into two types: securities, and OTC trades. The word derivative just means value derives from some reference source, and hence derivative can be either a security or an OTC trade.

CDOs and all the mortgage backed \”derivatives\” people keep talking about are not OTC products, and have nothing to do with the \”derivatives\” regulation the press talks about.

Posted by taxpayer | Report as abusive

Clearly we had misuses and abuses of the various OTC products in the last 20 years (just as firms and individuals time and time again got decimated by trading exchange listed derivatives). Clearly dealers made massive amounts of money on OTC derivatives sometimes ripping off clients (similar to the way insurance firms profited from writing insurance). The knee-jerk reaction regulation is however not the answer. Clearing platform solutions, standardization, and disclosure to regulators are sometimes helpful but represent only a fraction of the answer. The key is sound margining procedures (such as the ones used by exchanges) and proper bank capitalization. Most banks already have such procedures in place and the Fed (as well as the OCC) should review and push for strengthening of these practices. The industry is already moving in this direction. Destroying or restricting various risk management tools provided by dealers is unsound, even if it makes for great press coverage. If you want to understand instread of reacting, read a blog post entitled “Derivatives, the wrong war” at SoberLook.com


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/