Risk Markets And Politics

Thursday, December 14, 2006

Flora of North America (CFTC regulation again)

Flora vs. Iron Fence Tom Bell argues that the CFTC does not have authority over event and prediction markets. At the core of his argument is the definition of commodities subject to CFTC jurisdiction in the CEA: "all services, rights, and interests in which contracts for future delivery are presently or in the future dealt in." Bell claims that prediction markets offer "present delivery of conditional rights", not "future delivery of unconditional rights". Thus prediction markets are spot markets. However, this critical assertion hinges on whether "delivery" in the CEA definition above refers to delivery of rights. This is not the only possible reading, and I would bet that it is not the most common. Delivery could instead refer to delivery of a thing, or of cash, the amount of which is conditional. Bell also does not cite a relevant part of the CEA's definition of "excluded commodity": "an occurrence, extent of an occurrence, or contingency that is [...] associated with a financial, commercial, or economic consequence". This clearly could apply to an event or prediction market (though a market in science claims might not be considered "beyond the control of the parties to the relevant contract").

In any case, the CFTC has approved a few event markets with little challenge. With the caveat that I have no special legal expertise, I believe that the agency has legal jurisdiction over prediction markets, but does not want to exercise it for the reasons listed last time. Keep in mind that this whole legal and regulatory challenge is only relevant to exchange-traded retail contracts. If someone has a bright idea for a larger-scale commercial, industrial or governmental contract, there are plenty of options available, including an EBOT or OTC transactions. The "safe-harbor" suggestion of Paul Architzel might also entail stifling limitations. When the CFTC granted safe-harbor status to swaps in 1989, it did so under the conditions that the contracts: 1) have individually tailored terms, 2) be used in conjunction with the counterparty'¬ís line of business, 3) not be settled using exchange-style offset or a clearinghouse, and 4) not be marketed to the general public. Is it possible to negotiate a prediction market safe-harbor that would allow for some level of internet-based retail trading? Maybe, but what would it take? Likely more than a bunch of bloggers and scholars. In order to persuade the CFTC to spend time entertaining such a request, it would help greatly if you have a viable business ready to launch (and one which serves an uncontroversial social function). There is simply too much money at stake on the CFTC's other operations for them to devote resources to help someone disinterestedly sow seeds — even if the agency is completely sympathetic and agreeable in principle.

The no-action possibility seems to be off the table, especially if you want to operate an exchange for profit. On the gambling law front, it is encouraging that state prosecutors have left IEM alone for so long, but this is also probably a function of its non-profit status and lack of advertising.

All of that said, I continue to be optimistic about the possibilities of a real-money prediction exchange operating legally in the United States in the not-too-distant future. Finally, we all owe Tom Bell many thanks for the work he has done on this topic.

[Cross-posted from Midas Oracle]

0 Comments:

Post a Comment

<< Home