Bespoke Derivatives

Apparently commentators are upset because the administration's financial reform proposal only requires "standardized" derivatives to be cleared through central counterparties (CCPs), and not bespoke derivatives. James Kwak even questions "why we need customized derivatives in the first place." He claims he can't even think of an example of when a firm would legitimately need to use a bespoke derivative.

Umm, how about something like portfolio default swaps? A portfolio default swap is a credit derivative that transfers a predetermined amount of credit risk on a reference portfolio. For example, a pension fund with a large corporate bond portfolio might decide that it needs to reduce its exposure to 40 specific bonds that have a total notional amount of $100bn. The pension fund could enter into a portfolio default swap with a dealer in which the dealer agrees to protect the pension fund against the first $20bn of default-related losses on the reference portfolio (the 40 identified bonds) over the next 5 years, in exchange for quarterly premiums from the pension fund.

This kind of situation happens all the time, and it's perfectly legitimate for the pension fund to enter into the swap with the dealer. But portfolio default swaps can't be "standardized," because the reference portfolio in every swap is specific to the protection buyer. They're classic bespoke derivatives. Along the same lines, basket default swaps are also derivatives that can't be standardized. The most common basket default swap is a "first-to-default basket," in which payment is triggered by the first default—and only the first default—in a basket of identified bonds. Again, basket default swaps can't be standardized because the reference basket is specific to the protection buyer.

Also, CDS on structured products (e.g., ABS) are, by definition, bespoke derivatives because the reference ABS are all different. Yes, I know, CDS on CDOs are what brought down AIG, so some people think CDS on structured products are inherently evil. But that's not really a serious argument. The internet bubble brought down plenty of funds, but that doesn't mean stocks are inherently evil instruments. And anyway, the administration's proposal would impose conservative initial margin requirements on bespoke derivatives like CDS on structured products, which would prevent another institution from making a $400bn levered bet through OTC derivatives the way AIG did. I'm not even getting into classically bespoke derivatives like barriers (knock-in/knock-out options) or forward rate swaps, but suffice to say there are legitimate uses for all of these instruments.

I should also note that what Kwak claims is a "terrible example" of when a firm would need to use a bespoke derivative is, in reality, a very good example. The example was a firm that needs to hedge a credit exposure in an odd amount of money, and Kwak says that's a terrible example because the firm would be "perfectly fine" with only being able to buy CDS in multiples of $10,000. Spoken like someone who hasn't managed money a day in his life. (I actually laughed out lout when I read that part.) I don't know a single corporate treasurer or portfolio manager who would be "perfectly fine" with that. Hedging in odd amounts of money is a very common—and perfectly legitimate—use of bespoke derivatives.