How to Fix Wall Street – Madoff’s Law

This is an absolutely terrible idea. A couple problems:

1) It creates huge concentrations of counterparty risk. Let's say you're the firm that invents the credit default swap. For the next five years, you are the counterparty for every single CDS transaction. If you go bust, the whole market explodes and everyone is in for a world of hurt. Participants can't spread their risk around by trading with different counterparties. In the current environment, we should be doing everything we can to decrease counterparty risk.

2) What counts as a "new" financial instrument? Let's say it's 2004 and you want to start selling credit default swaps on asset-backed securities. At that point, CDS had been around for 10 years and ABS for at least 15. So is CDS on ABS a new product? Can you get a patent for it? For another example, let's say it's the mid-80s and you work on the CMO desk. (CMOs, collateralized mortgage obligations, are a type of mortgage-backed-security where the cashflows from the mortgages are restructured to better fit investors' demands). Everyone's doing sequential deals, and you do the first floater/inverse IO deal. CMOs aren't a new product, but your structure is completely original. Can you get a patent on that?

3) The products that are at the core of the current crisis weren't really new! Mortgage backed securities have been around in one form or another since the late 70s, and credit default swaps were invented in the early 90s. Any patents would have long since expired.

4) You'd be handing huge profit margins to the bank that invents them. For traditional physical products, the idea is that this government-granted monopoly encourages companies to spend time doing capital-intensive R&D which they wouldn't otherwise consider. But we've seen that banks, hedge funds, etc. have been willing to innovate without government-granted monopolies. So why give them a monopoly now?

Ultimately, this isn't much different from software patents. If software consists of an abstract idea instantiated in code, financial instruments are an abstract idea instantiated in legal documents and markets. The case against software patents applies to financial instruments too.

This ties into what I like to call the "other gambler's fallacy", wherein one's visceral understanding of probability tends to exaggerate high probabilities and underestimate low probabilities. That is, one sees a probability like 0.001, and ends up treating it as though it were zero.

That said, I'm not sure about the solution the guy proposes. It's basically patents for esoteric financial instruments. While I don't see anything inherently wrong with the idea, patents as they are now nominally exist to speed innovation. This idea is intended to slow the spread of innovations in one particular area.

In other words, it's possible that this system could develop all the problems that the patent system has: people would come up with slight variations on an existing financial instrument so that they could sell it without buying a license from the creator, or for marketing reasons, or whatever. This would naturally make instruments more complicated, which in turn makes them harder to reason about.

Worthwhile part:

>Call it Madoff’s Law. Where there are consistent returns, it doesn’t mean that the risk is reduced, it means you can’t see the risk.

Worthless part:

>I would create additional forms of licenses (...) that any financial institution can apply for (...) and receive for a new financial instrument.

If this extra cost/regulation was enacted it would just move the business offshore.