Smart or Not So Smart Money; The Limits on Derivatives and Regulating Them
posted by Deven Desai
The New York Times op-ed by Calvin Trillin, Wall Street Smarts, has a parable-like quality with the two characters meeting and exchanging wisdom. The lesson offered by the wiseman: “The financial system nearly collapsed,” he said, “because smart guys had started working on Wall Street.” The piece goes on to explain why that is a good explanation. It seems that the not-so-smart sat at the top of the heap and ran the companies: “Guys who didn’t have the foggiest notion of what a credit default swap was. All our guys knew was that they were getting disgustingly rich, and they had gotten to like that.” There is also an claim about what is enough and what is greed in this tale. I leave it to others to debate or verify these ideas (our own Mr. Cunningham has been a favorite for me on these issues). Now, a paper by some folks at Princeton may show that not even the smart guys knew what they were doing.
As Andrew Appel explores in his post Intractability of Financial Derivatives, the computer science world’s Intractability Theory may better explain the derivative world than other theories. (the theory is used for DRM, cryptography, and more). The paper is Computational Complexity and Information Asymmetry in Financial Products (pdf) by Sanjeev Arora, Boaz Barak, Markus Brunnermeier, and Rong Ge.
For those who are interested in the topic and/or understand the math and theory behind the risk shifting involved in this area, check out Andrew’s post. He does a great job explaining how the paper applies to a CDO (collateralized debt obligation). If you need a little more to understand why this paper and its ideas are important, consider Andrew’s take away
In principle, an alert buyer can detect tampering even if he doesn’t know which asset classes are the lemons: he simply examines all 1000 CDOs and looks for a suspicious overrepresentation of some of the asset classes in some of the CDOs. What Arora et al. show is that is an NP-complete problem (“densest subgraph”). This problem is believed to be computationally intractable; thus, even the most alert buyer can’t have enough computational power to do the analysis.
Arora et al. show it’s even worse than that: even after the buyer has lost a lot of money (because enough mortgages defaulted to devalue his “senior tranche”), he can’t prove that that tampering occurred: he can’t prove that the distribution of lemons wasn’t random. This makes it hard to get recourse in court; it also makes it hard to regulate CDOs.
UPDATE: It appears from the comments to Andrew’s post that CDO and derivatives are not precisely the same thing. In addition, the comments explore the limits of the study. It is a good discussion.
ALSO check out the FAQ for the paper. It addresses many issues that the initiated may want to probe.
October 18, 2009 at 9:17 am
Tags: computer science, derivatives, securities law
Posted in: Corporate Finance, Corporate Law, Securities, Securities Regulation
Print This Post
Responses (4)
Mike Zimmer - October 19, 2009 at 7:54 am
In the comments to the linked article, a commentator describes credit default swaps this way: “you and I could contract for payments between each other depending on almost any third referent, exchange rates coffee prices etc. Neither you nor I have to actually own any foreign currency or coffee however.”
As I understand, these type contracts are free of federal regulation, accomplished by Wall Street during the Clinton era. But, isn’t this a gambling contract? How is this different than you and me contract on the third referrant — the outcome of the Monday Night NFL game — over which neither of us have any direct interest? Wouldn’t normal state anti-gambling laws make this illegal and thus unenforceable? Did Congress, in freeing this stuff from federal regulation, also preempt state anti-gambling laws?
Nate Oman - October 22, 2009 at 8:06 am
Deven: a CDO is a kind of derivative, but not all derivatives are CDOs. A CDO entitles its holder to a fractional share in the income stream generated by a pool of financial assets. It is thus a derivative in that the value of the CDO is dependent on, i.e. derivative of, some underlying financial asset. On the other hand, a CDS is also a derivative. In this case A promises to make a payment to B if there is a shift in the value of some referenced financial asset. The value of the CDS is dependent on, i.e. derivative of, the referenced financial asset. On the other hand, the payouts under a CDS do not come directly from the financial asset as in the case of a CDO.
Mike: If a CDS is a gambling contract, so are virtually all commodity futures contracts, which also consist of cash payments based on market price flucuations in an underlying asset which needn’t be owned by either part to the contract. Indeed, if you know anything about the history of the law of contracts treatment of futures contracts in the 19th century, there is a weird feeling of deja vu in the rhetoric surrounding CDSs. One can use a futures contract to simply place a bet. One can also use a futures contract to hedge against risk. Today, we don’t try to draw a distinction between which way a party is using the future contract. Rather than condemning the contract as a gambling device, we create regulatory mechanisms — essentially clearing house rules — to deal with counter-party risk, short-squeeze risk and the like. Solving these problems is not rocket science. It seems to me that problem with CDS is not in the contracts themselves, but in the regulatory and — far more importantly importantly — the political and monetary environment in which institutions could play in the CDS market. The culprit here is not a liquid market in credit risk. Rather, the culprits are lax capital requirements, loose monetary policy, and a bail-out political culture with huge moral hazards.
Mike Zimmer - October 27, 2009 at 8:19 am
Thanks, Nate, for your clear explanation.
Mike Zimmer - October 27, 2009 at 8:21 am
Thanks, Nate
Leave a Reply