Would you pay good money for accurate predictions about important events, such as election results or military campaigns? Not if the U.S. Commodity Futures Trading Commission (CFTC) has its way. It recently took enforcement action against overseas prediction markets run by InTrade and TEN. The alleged offense? Allowing Americans to trade on claims about future events.
The blunt version: If you want to put your money where your mouth is, the CFTC wants to shut you up.
A prediction market allows its participants to buy and sell claims payable upon the occurrence of some future event, such as an election or Supreme Court opinion. Because they align incentives with accuracy and tap the wisdom of crowds, prediction markets offer useful information about future events. InTrade, for instance, accurately called the recent U.S. presidential vote in all but one state.
As far as the CFTC is concerned, people buying and selling claims about political futures deserve the same treatment as people buying and selling claims about pork futures: Heavy regulations, enforcement actions, and bans. Co-authors Josh Blackman, Miriam A. Cherry, and I described in this recent op-ed why the CFTC’s animosity to prediction markets threatens the First Amendment.
The CFTC has already managed to scare would-be entrepreneurs away from trying to run real-money prediction markets in the U.S. Now it threatens overseas markets. With luck, the Internet will render the CFTC's censorship futile, saving the marketplace in ideas from the politics of ignorance.
Why take chances, though? I suggest two policies to protect prediction markets and the honest talk they host. First, the CFTC should implement the policies described in the jointly authored Comment on CFTC Concept Release on the Appropriate Regulatory Treatment of Event Contracts, July 6, 2008. (Aside to CFTC: Your web-based copy appears to have disappeared. Ask me for a copy.)
Second, real-money public prediction markets should make clear that they fall outside the CFTC's jurisdiction by deploying notices, setting up independent contractor relations with traders, and dealing in negotiable conditional notes. For details, see these papers starting with this one.
[Aside to Jerry and Adam: Per my promise.]
[Crossposted at Technology Liberation Front, and Agoraphilia.]
Wednesday, December 12, 2012
CFTC Targets Prediction Markets; Hits First Amendment
Saturday, April 30, 2011
U.S. Supreme Court Prediction Market
Recently posted to SSRN: FantasySCOTUS: Crowdsourcing a Prediction Market for the Supreme Court, a draft paper by Josh Blackman, Adam Aft, & Corey Carpenter assessing the accuracy of the Harlan Institute's U.S. Supreme Court prediction market, FantasySCOTUS.org. The paper compares and contrasts the accuracy of FantasySCOTUS, which relied on a "wisdom of the crowd" approach, with the Supreme Court Forecasting Project, which relied on a computer model of Supreme Court decision making. From the paper's abstract:
During the October 2009 Supreme Court term, the 5,000 members made over 11,000 predictions for all 81 cases decided. Based on this data, FantasySCOTUS accurately predicted a majority of the cases, and the top-ranked experts predicted over 75% of the cases correctly. With this combined knowledge, we can now have a method to determine with a degree of certainty how the Justices will decide cases before they do. . . . During the October 2002 Term, the [FantasySCOTUS] Project’s model predicted 75% of the cases correctly, which was more accurate than the [Supreme Court] Forecasting Project’s experts, who only predicted 59.1% of the cases correctly. The FantasySCOTUS experts predicted 64.7% of the cases correctly, surpassing the Forecasting Project’s Experts, though the difference was not statistically significant. The Gold, Silver, and Bronze medalists in FantasySCOTUS scored staggering accuracy rates of 80%, 75% and 72% respectively (an average of 75.7%). The FantasySCOTUS top three experts not only outperformed the Forecasting Project’s experts, but they also slightly outperformed the Project’s model - 75.7% compared with 75%.
You can download a copy of the draft paper here.
[Crossposted at Agoraphilia, Midas Oracle, and MoneyLaw.]
Monday, July 07, 2008
Free Speech in Event Market Claims
In addition to a joint comment with a score of signers, I also responded to the Commodity Futures Trading Commission (CFTC)'s Concept Release on the Appropriate Regulatory Treatment of Event Contracts by firing off a solo comment. I there focused soley on question 14, in which the CFTC asked, "Should certain underlying events or measures--such as those based on assassinations or terrorist activities—be prohibited altogether due to the social perception and impact of such events? What statutory or other legal basis would support this treatment?"
Much of my comment tracked the answer I posted here earlier. Long story short: such claims would not materially promote wrongful acts, so the CFTC has no legal basis to ban them. To that argument, however, I added a First Amendment analysis; to wit:
Could you fault claims about assassinations or other terrorist events for giving incentives for wrongful acts? Not very plausibly; as I explain above, it is very unlikely that anyone would find it profitable or prudent to try to use an event market to cash in on wrongdoing. Furthermore, all sorts of investment instruments offer the same incentives. Thus, for instance, a would-be terrorist might go long on oil futures prior to pulling off an attack on a refinery. Indeed, that sort of scenario seems much, much more likely than one involving event markets.
At root, concern about unseemly event market claims boils down to concern about violating a taboo about what sorts of things people discuss openly in a polite society. Those norms merit our concern, granted. They do not, however, justify imposing a speech restriction on event markets. And make no mistake about it; to bar such claims would constitute a restriction on speech.
Specifically, if the CFTC banned certain sorts of event market claims relating to assassinations, terrorist activities, or criminal acts, it would thereby impose a content-based restriction on speech. That would, under present First Amendment jurisprudence, trigger the highest level of judicial review: strict scrutiny. The ban would almost certainly fail to survive that scrutiny, as it would be too broad (stopping not just the bad guys but also the good ones), too narrow (since it would fail to forbid the use of other financial instruments, such as generic futures, from like uses), and not narrowly tailored (since there are other, better ways to discourage bad acts). Those sorts of claims would, moreover, fall within the core of the sort of speech protected by the First Amendment, as they would concern political events.
Our freedoms of speech and expression include the right to ask troubling questions. The CFTC has no good reason to ban event market claims about assassinations or other illegal acts. Nor can it do so constitutionally.
Suppose that my argument leads the CFTC to doubt that it can ban event markets from hosting claims about assassinations or terrorist activities. Suppose further, as seems likely, that the CFTC has some discretion in deciding whether its jurisdiction encompasses event markets; suppose, that is, that extant law does not command one answer to that question. What result?
I predict that the CFTC would tend to deny that it has jurisdiction over event markets because it would not want to take the blame for encouraging distasteful claims. Indeed, I not only predict that result, I intend it. I don't trust the CFTC to do a very good job regulating event markets, so I want it to know why it does not even want to try.
[Crossposted at Agoraphilia and Midas Oracle.]
The CFTC Deadline . . . Wavers
The Commodity Futures Trading Commission (CFTC)'s Concept Release on the Appropriate Regulatory Treatment of Event Contracts says, "Comments must be received by July 7, 2008." What deadline does that impose? I played it safe, and assumed that I had to send mine in before midnight, this morning. Today, I learned from Bruce Fekrat, Special Counsel to the Office of the Director of the CFTC, that in fact I had up until this coming midnight to email my comments.
Any lawyer worth his or her salt knows better than to use the ambiguous "by" when specifying a deadline. The CFTC should have said, "Comments must be received before July 8, 2008." Thanks to its poor drafting, I rushed my comment out the door, missed the chance to get lots of new signatures, and wasted time swapping emails with various people equally confounded about what the CFTC meant.
Sigh. And to think that some people hope the CFTC will assert jurisdiction over prediction markets and clarify their legality under U.S. law! If the CFTC cannot get even something so routine as specifying a deadline right, I shudder to think what sort of confusion it might impose on prediction markets.
[Crossposted at Agoraphilia and Midas Oracle.]
Thursday, July 03, 2008
Let's Tell the CFTC Where to Go
Update: I've extended the deadline for signing up until 7 p.m. Pacific, Sunday, July 6. Also, I fixed a typo in paragraph 3, changing "denying" to "giving." (Thanks, Gil!)
The deadline looms for interested parties to respond to the Commodity Futures Trading Commission's request for comments about regulating prediction markets ("event markets" in the CFTC's usage). I may or may not get around to a detailed, point-by-point response to the CFTC's many questions. In the meantime, though, I've drafted a general statement that many of you might agree with. I invite you to sign it with me, so that together we might tell the CFTC where to go. Please see below for details on how to sign on. Here is the draft statement:
What regulatory treatment should the Commodities Futures Trading Commission ("CFTC") apply to event markets? We the undersigned, who represent a wide range of viewpoints, agree on three general observations. First and foremost, the CFTC should do no harm. Second, at a minimum, the CFTC should make more general the sort of "no action" status enjoyed by the Iowa Electronic Markets ("IEM"). Third, if the CFTC decides to regulate event markets more substantively, it should adopt clear and limited jurisdictional boundaries and allow affected parties to step outside of them.
First, do no harm: Many sorts of event markets—including public ones, private ones, ones that offer only play-money trading, and ones that offer real-money trading—already thrive in the U.S. They have provided a rich array of benefits without evidently harming anyone. The CFTC could help event markets achieve still greater success by clarifying their legality. Instituting the wrong sort of regulations could suffocate event markets in their cradle, however. The CFTC should exercise a light hand, taking care to do no more than offer qualifying event markets the shelter of federal preemption and freeing them to continue operating under the extant legal regime.
Second, open up the "no action" option: Thanks in part to the "no action" letters that the CFTC has issued to it, the IEM has for many years benefited the public by offering real-money event markets. No sound reason precludes the CFTC from giving similar treatment to other institutions that, like the IEM, offer event markets solely for academic and experimental purposes and without imposing trading commissions.
Although the CFTC's "no action" letters do not specify the exact criteria the IEM had to satisfy, they took favorable note of the IEM's account limits. Those account limits effectively prevent the IEM from supporting significant hedging functions. If the CFTC builds a similar requirement into any general "no action" guidelines, it should adopt limits considerably more generous than the meager $500/trader limit adopted decades ago by the IEM. Even a limit ten times that amount would still effectively preclude hedging.
The CFTC should not limit "no action" status to markets run by tax-exempt organizations. The no-action letters that the CFTC issued to the IEM emphasized not the nature of the hosting institution, the University of Iowa, but rather the business model adopted by the IEM itself. Profitability could not have mattered, as tax-exempt organizations can and do earn profits (indeed, as their burgeoning endowments demonstrate, many universities earn immense profits). The CFTC apparently cared only that the IEM did not plan to profit from charging traders commissions. A tax-paying organization could satisfy that condition just as easily as a tax-exempt organization could. In either event, price discovery would flourish and consumers would win a safeguard against getting fleeced.
Third, preserve regulatory exit options: If the CFTC decides to write substantive regulations for event markets, it should recognize and guard against the risk of overregulation. Even well-intentioned and well-informed regulators remain human and, thus, all too apt to make mistakes. They run an especially large risk of making mistakes when they first attempt to regulate new institutions, such as event markets. To make matters worse, regulators typically lack reliable signals to determine when they have gone too far. Industries wither away for many reasons, after all.
The CFTC's approach to regulating event markets should accommodate these policy considerations by establishing clear jurisdictional boundaries and preserving regulatory exit options. Thus, for instance, the CFTC might specify that it has no jurisdiction over event markets that offer trading only to members of a particular firm, over markets that offer only spot trading in negotiable conditional notes, or over markets that do not support significant hedging functions.
The CFTC's approach to regulating event markets should accommodate these policy considerations by establishing clear jurisdictional boundaries and opening exit options. Thus, for instance, the CFTC might specify that it has no jurisdiction over event markets that offer trading only to members of a particular firm, over markets that offer only spot trading in negotiable conditional notes, or over markets that do not support significant hedging functions. Then, if the CFTC enacts unduly burdensome regulations, an event market could opt out of them by changing its business model. So long as markets publicly announce that they operate outside the CFTC's purview, allowing them that freedom of exit would harm nobody. To the contrary, it would help the CFTC gauge the suitability of its regulations and serve the public by protecting the continued viability of event markets.
Interested in signing on? Please drop me a private email (tbell at chapman dot edu) with your name, institutional affiliation, and snailmail address. I welcome your comments—I'm sure a typo or two persists in my draft—but I of course cannot revamp the entire statement without mucking up the entire process. To leave me time to get everything together and out the door before the July 7 deadline, you'll have to contact me before noon Pacific time on Sunday, July 6.
[Crossposted at Agoraphilia and Midas Oracle.]
Monday, June 23, 2008
Let Prediction Markets Fight Terrorism
The Commodity Futures Trading Commission (CFTC)'s recent request for comments about the regulation of prediction markets includes a number of specific questions. I am not sure whether I will manage to write up answers to all of them before the July 7 deadline, but question in particular—question 14—has attracted my attention. The CFTC there asks, "Should certain underlying events or measures--such as those based on assassinations or terrorist activities—be prohibited altogether due to the social perception and impact of such events? What statutory or other legal basis would support this treatment?"
I answer the first part of question 14, "No," (and thus need not answer the second part). I doubt that the CFTC wants to hear that sort of reply, frankly; I instead suspect that it wants a legal excuse to avoid the sort of political firestorm that followed the Pentagon's proposal to create a Policy Analysis Market that included claims about assassinations and terrorist events. My draft answer to question 14 explains why I'm willing to risk disappointing the CFTC:
The CFTC should not forbid trading in claims based on assassinations, terrorist activities, or other criminal acts. Because event markets would offer only relatively thin and traceable trading, they would not offer an attractive investment option to anybody planning to profit from wrongdoing. A would-be terrorist would risk revealing both his plans and his identity if, for instance, he invested in a contract predicting another 9/11. He would instead find it more safe and profitable to simply short certain publicly traded stocks.
Furthermore, event markets in terrorist or criminal acts might benefit the public by revealing life-saving information. Suppose, for instance, that an anthropologist's study of corrido culture convinced her that narcoterrorists had begun planning military raids on border checkpoints in Arizona and California. If she had the opportunity to buy terrorist event claims, she might both profit from her research and tip us all off about looming trouble. Sound public policy suggests that we should encourage that sort of trading—not forbid it.
To judge from their reactions to the Policy Analysis Market proposed by the Pentagon in 2003, politicians might need to learn more about the benefits of using trading to help predict assassinations or other terrorist events. That poses a public relations problem, however—not a legal one. The CFTC thus has no sound reason to presumptively forbid trading in contracts related to such events.
Notably, my answer to question 14 differs sharply from the answer offered by Jed Christiansen. He said, "There should never be any incentive to break a law, so there should never be any contracts that would pay someone if a law was broken." I disagree, of course, but I thank him for stimulating me to offer an alternative take.
[Crossposted at Agoraphilia and Midas Oracle.]
Monday, May 19, 2008
Building Exits into CFTC Regulation
Much of my draft paper, Private Prediction Markets and the Law, focuses on nuts-and-bolts fixes for the legal uncertainty that currently afflicts private prediction markets under U.S. law. I'll say more about those in later posts to Agoraphilia and Midas Oracle. The paper also dicusses a more theoretical and general issue, though: The benefits of designing regulatory schemes to include exit options.
The Commodity Futures Trading Commission recently issued a request for comments about whether and how it should regulate prediction markets. In earlier papers, I explained why the CFTC cannot rightly claim jurisdiction over many types of prediction markets. I recap that view in my most recent paper, but add some suggestions about how the CFTC might properly regulate some types of prediction markets. In brief, I suggest that the CFTC build exit options into any regulations it writes for prediction markets, allowing those who run such markets the same sort of freedom of choice that U.S. consumers already enjoy, thanks to internet access to overseas markets like Intrade, with regard to using prediction markets. Here's an excerpt from the paper:
Those practical limits on the CFTC's power should encourage it to write any new regulations so as to allow qualifying prediction markets to operate legally, and fairly freely, under U.S. law. . . . Ideally, the CFTC would offer prediction markets something like these three tiers, each divided from the next with clear boundaries.
- Designated Contract Markets. Regulations designed for designated contract markets, such as the HedgeStreet Exchange, would apply to retail prediction markets that offer trading in binary option contracts and significant hedging functions.
- Exempt Markets. Regulations for "exempt" markets, which impose only limited anti-fraud and manipulation rules, would apply to prediction markets that:
- offer trading in binary option contracts;
- thanks to market capitalization limits or other CFTC-defined safe harbor provisions do not primarily support significant hedging functions; and
- offer retail trading on a for-profit basis.
- No Action Markets. A general "no action" classification, similar to the one now enjoyed by the Iowa Electronic Markets, would apply to any market that duly notifies traders of its legal status and that is either:
- a public prediction market run by a tax-exempt organization offering trading in binary option contracts but not offering significant hedging functions;
- a private prediction market offering trading in binary option contracts, but not significant hedging functions, only to members of a particular firm; or
- any prediction market that offers only spot trading in conditional negotiable notes.
Notably, regulation under either of the first two regimes would definitely afford a prediction market the benefit of the CFTC's power to preempt state laws. It remains rather less clear whether the third and lightest regulatory regime would offer the same protection, though the cover afforded by its two "no action" letters has allowed the Iowa Electronic Markets to fend off state regulators. Markets that by default qualify for the third regulatory tier described above thus might want to opt into the second tier, so as to win a guarantee against state anti-gambling laws and the like. So long as they satisfy the first two conditions for such an "exempt market" status, public prediction markets run by non-profit organizations or private prediction markets that offer trading only to members of a particular firm should have that right. Why offer this sort of domestic exit option? Because it would, like the exit option already open to U.S. residents who opt to trade on overseas prediction markets, have the salutatory effect of curbing the CFTC's regulatory zeal.
The footnotes omitted from the above text includes this observation: "Because they fall outside the CFTC's jurisdiction, markets offering only spot trading in conditional negotiable notes could not opt into the second regulatory tier."
Please feel free to download the draft paper and offer me your coments.
[Crossposted at Agoraphilia, Technology Liberation Front, and Midas Oracle.]
Wednesday, May 07, 2008
Insider Trading and Private Prediction Markets
People who run in-house, corporate prediction markets have told me that U.S. laws against illegal insider trading give them nightmares. The problem arises because a private prediction market typically generates material nonpublic information about the corporation that hosts it. If somebody misuses that information to time the purchase or sale of the corporation's stock, liability for illegal insider trading could follow.
I plan to say a great deal more about this problem, and some proposed cures, in a paper I'm writing for the Journal of Prediction Markets. In very brief, I propose several strategies that should help to mitigate the risks that private prediction markets create under illegal insider trading laws:
- Segregate markets for traditional insiders from other markets.
- Broaden safeguards against illegal insider trading to reach beyond traditional insiders.
- Treat the market's claims and prices as trade secrets.
- Set up decoy claims and prices.
Alternatively, of course, a corporation could simply make public the claims and prices of its in-house prediction market. Illegal insider trading laws only speak to material nonpublic information, after all. It seems very unlikely that any corporation would willing disclose so much and such probative information about its management, however.
[Crossposted at Agoraphilia and Midas Oracle.]
Thursday, February 28, 2008
Getting from Collective Intelligence to Collective Action
I really enjoyed attending the Collective Intelligence FOO Camp, sponsored by Google and O'Reilly Media, last weekend. I'd been expecting a sort of geek slumber party, and had looked forward to rolling out my awesome Darth Vader impersonation. I was all set to cut loose with a growling, "I'm your father, Luke." It didn't quite come to that, but I still had a blast, meeting lots of smart, informed, articulate, creative, and successful people. Friendly people, too.
I described how to establish the legality of real money, open-access prediction markets under U.S. law. I called my presentation, Getting from Collective Intelligence to Collective Action [PPT file]. In very brief, I proposed this algorithm:
- Set up an enterprise prediction market, make playing it a condition of continued employment, and offer valuable prizes to the best predictors.
- Set up a limited access prediction market, hire a number of independent contractors researchers to play it, pay them a relatively low salary for doing so, and offer valuable prizes to the best predictors.
- Set up an open-access prediction market but require anyone playing it to go through a click-wrap license that creates the sort of independent researcher relationship described at step 2, above.
When and if standing for declaratory judgment obtains, a litigation team should bring suit seeking to establish the legality of the prediction market under U.S. law. The market should be run by a worthy institution and deal only in claims likely to generate large positive externalities. Google.org, for example, might set up a market in earthquake claims and ask for a court's blessing.
That strategy would stand a fair chance—a 75% chance, I'd say—of establishing the legality of a great many in-house and (effectively) public prediction markets under U.S. law. The strategy would not impose great costs or risks, though it would take some careful planning and execution, and would almost certainly generate large private and public goods. It might even save lives; we could really use a reliable early-warning system for major earthquakes.
This legalization program would most directly benefit subsidized markets; it would not plainly establish the legality of markets where traders could invest their own funds or hedge against off-market risks. I'm still working out a legal hack for that step. It will be an easier step to take, however, if we've already made it as far as establishing the legality of open-access, real money prediction markets under U.S. law.
We should still pursue other routes to establishing the legality of prediction markets under U.S. law, of course. It already helps that so many U.S. residents evidently make and lose money on Ireland-based InTrade. That at least goes to show that prediction markets hurt nobody. Getting the CFTC to issue safe-harbor regulations would help, too. All of those routes to public, real money prediction markets in the U.S. merit exploration. Any of them might offer a shortcut to a better, freer future.
[Posted at Agoraphilia, The Technology Liberation Front, and Midas Oracle.]
Thursday, February 21, 2008
Quake Markets
Markets offer us a potentially useful tool for predicting earthquakes. Imagine the San Andreas fault divided into segments, each of which carries a price based on the present discounted disvalue of a future quake. That price would reflect both a quake's place in time and its place on the Richter scale. Such a market in quake claims would probably generate some useful--even lifesaving--data. If sufficiently thick, it might offer hedging, too.
I've not yet found that sort of quake market. Has any of you? If none exists, at least one should! Plenty of people and institutions would love to know more about earthquakes. Some of them would gladly support an earthquake market, I'd bet. There remain some legal risks, granted, but I think I've got a good hack for those. (Long story short: independent contractor researchers paid a base salary for making trades and winning bonuses for correct predictions.)
One nice thing about a quake market: Done right, it would generate powerfully positive externalities, benefiting even those who do not trade on the market. Imagine a map of the San Andreas fault, the price of each tradable segment illustrated by color coding or line thickness. One glance at that picture, and you would know whether it was time to relax, double-check your emergency kit, or head for the hills.
[Crossposted to Midas Oracle.]
Wednesday, October 31, 2007
Presentation of Private Prediction Markets’ Legality Under U.S. Law
Koleman Strumpf has tapped me to discuss the legality of in-house prediction markets, tomorrow, at the Conference on Corporate Applications of Prediction/Information Markets, at the Kauffman Foundation Conference Center, in lovely Kansas City, MO. I call my presentation, Private Prediction Markets’ Legality Under U.S. Law. It discusses the legal hurdles faced by in-house PMs, and some possible cures. You can download the PowerPoint file, together with ample notes, here. I look forward to meeting many fellow PM-geeks, and to hearing the thoughts of Robert E. Litan and the other conference participants.
[Crossposted to Midas Oracle]
Wednesday, June 27, 2007
Paper on Corporate Prediction Markets
Although it's now a few years old, I only recently read this fine paper about how to build useful corporate prediction markets: Martin Spann & Bernd Skiera, Internet-Based Virtual Stock Markets for Business Forecasting, 49 Management Science 1310 (2003). It offers a very accessible and comprehensive discussion of why and how on-line prediction markets can help businesses predict short-to-medium-term events. You'll find the paper well worth reading even if you skip over the few mathematical bits of the text.
[Crossposted to Midas Oracle.]
Monday, March 05, 2007
Do Gambling Laws Threaten Prediction Markets?
Because prediction market traders generally rely more on skill than on chance to win, I've argued that they do not qualify in "gambling" as defined by U.S. law. I've also admitted, however, that over-ambitious prosecutors might see the matter differently. A recent email exchange I had with Prof. George R. Neumann, who serves on the Board of Directors of the Iowa Electronic Markets, confirmed my worries. I here quote him, with his permission.
"We have been threatened several times with suits by various states," he wrote, "but so far the CFTC coverage [i.e., the two 'no action' letters that the CFTC gave the IEM] has been our trump card." Prof. Neumann offered this example: When the IEM ran a market on Hilary Clinton's run for the New York Senate seat, an N.Y.C.-based district attorney "contacted us (via a very nasty letter) to tell us that NY had a law that forbade gambling on elections in that state. He demanded a list of each and every NY State resident who was a participant in our market." Wielding the CFTC's "no action" letter in defense of the IEM, Prof. Neumann directed the DA to file suit in, um, a venue at once very local and yet very inconvenient.
Alas for those who would like to match Neumann's feat of legal acumen, the CFTC's treatment of the IEM so far looks like a one-shot deal. Or, rather, it looks like a big-shot deal; you would probably need some political help to squeeze another such letter out of the CFTC. While you're at it, please ask the CFTC to issue an advisory opinion saying, "The CFTC has no rationale to regulate transactions or markets in which price discovery functions predominate over hedging functions." It could state a market cap limit to make clear the limits on its discretion.
Such a "public no action letter" would give a great many useful prediction markets freedom to help us discern the future, just as the IEM has done. It would also, to judge from the IEM's experiences, give new prediction markets a shield from state anti-gaming prosecutions.
[Crossposted to Midas Oracle.]
Tuesday, February 20, 2007
Presentation on Legality of Prediction Markets Under U.S. Law
On Thursday, at the SoCal Collective Intelligence Networks Summit, I'm slated to talk about the legality of prediction markets under U.S. law. You can here download the PowerPoint file I plan to use. It offers some general observations about stategies to make real-money prediction markets conform with U.S. law. For details, I welcome you to attend the talk. Please introduce yourself as an Agoraphilia reader if we meet!
[Crossposted to Midas Oracle.]
Saturday, December 16, 2006
Prediction Markets "Excluded" from CFTC Jurisdiction?
Can the Commodity Futures Trading Commission ("CFTC") properly claim jurisdiction over prediction markets? Jason Ruspini recently tackled that difficult question. Following a gracious caveat about the limits of his legal expertise, Jason concluded that "the agency has legal jurisdiction over prediction markets, but does not want to exercise it" for various reasons.
I've addressed the same question at some length in Gambling for the Good, Trading for the Future: The Legality of Markets in Science Claims, 5 Chapman L. Rev. 159, 172-76 (2002), and more briefly in Prediction Markets for Promoting the Progress of Science and the Useful Arts, 14 Geo. Mason L. Rev. __ (2006) (forthcoming). Jason refers only to the later paper, which relies on the earlier one for detailed analysis. That perhaps explains his complaint that, "Bell [] does not cite a relevant part of the [Commodity Exchange Act's] definition of 'excluded commodity' . . . ." I do cover that topic; I simply don't do so in the paper that Jason relies on.
What do I say in that earlier paper? First off, consistent with my later paper, I argue that the CEA does not give the CFTC jurisdiction over run-of-the-mill prediction markets. "The CEA does not cover contracts intended to effectuate future delivery, much less contracts that effectuate immediate delivery." 5 Chapman L. Rev. at 170-71 (footnote omitted). With that in mind, it becomes obvious why prediction markets should generally fall outside the CFTC's jurisdiction. Such markets typically deal in the present delivery of conditional rights: claim coupons payable in the event some contested claim comes true.
As all my writings on the topic have taken pains to concede, however, defining the exact scope of the CFTC's jurisdiction proves frustratingly difficult. A court might well decide that my argument smacks of logic-chopping. The Gambling for the Good paper thus spends a fair amount of time exploring what would transpire if the CFTC did win jurisdiction over prediction markets. Specifically, pages 172-76 discusses whether the claims traded on a prediction market would qualify as "excluded commodities" and, if so, what sort of regulations would apply. Here's my bottom line:
[S]hould a market in science claims find itself subject to the CEA, it could attempt to qualify for the “excluded commodity” loophole that would largely free it from CFTC regulation. It looks highly probable, however, that the CFTC would have wide discretion to thwart any such attempt. At the least, to judge from precedent, the CFTC would probably not exclude a market in science claims from its regulations without also imposing crippling conditions.
Friends of prediction markets should not despair, however. First, as I've said, there remain good reasons to deny the CFTC jurisdiction over prediction markets. Second, the "hybrid instrument predominantly a security" category, described in Prediction Markets for Promoting the Progress of Science, offers yet another promising but untried legal hack for both preempting state gaming laws and forestalling CFTC regulation. Third, various state law pertaining to games of skill may well provide shelter for prediction markets.
That last legal strategy remains, so far as I can tell, completely unexplored in the extant literature on prediction markets. My initial research on that topic has proved heartening, encouraging me to learn more. I'll plan to say more about the topic later.
[Crossposted to Midas Oracle.]
Thursday, October 05, 2006
Safe Harbor for Prediction Markets in SAFE Port Act
Congress recently passed a law, the Unlawful Internet Gambling Enforcement Act of 2006, that aims to restrict access to certain forms of online gaming. It has not, contrary to some reports, been passed into law. The House and Senate passed the UnInGEn-ious Act (as I style it) as subchapter IV of the SAFE Port Act, H.R. 4954, 109th Cong., 2nd Sess. (2006), which as yet lacks a presidential signature. Let's assume that it will get one, though. What will the enactment of the SAFE Port Act, and thus of the UnInGEn-ious Act, portend for prediction markets? Nothing very scary.
Appropriately, given the title of the larger bill into which it got tucked, the UnInGEn-ious Act creates a safe harbor that prediction markets can probably take shelter in. The Act targets only "unlawful internet gambling," which it defines as the knowing transmission of "a bet or wager by any means which involves the use, at least in part, of the Internet where such bet or wager is unlawful under any applicable Federal or State law . . . ." Section 5362(10)(A). If prediction markets do not involve bets or wagers, therefore, they escape the reach of the UnInGEn-ious Act.
Helpfully, the Act defines "bet or wager" to not include
participation in any . . . educational game or contest . . . that meets the following conditions:Id. section (1)(E)(ix).(I) All prizes and awards offered to winning participants are established and made known to the participants in advance of the game or contest and their value is not determined by the number of participants or the amount of any fees paid by those participants.
(II) All winning outcomes reflect the relative knowledge and skill of the participants and are determined predominantly by accumulated statistical results of the performance of individuals (athletes in the case of sports events) in multiple real-world sporting or other events.
(III) No winning outcome is based--(aa) on the score, point-spread, or any performance or performances of any single real-world team or any combination of such teams; or
(bb) solely on any single performance of an individual athlete in any single real-world sporting or other event.
Does that definition of "bet or wager" exclude prediction markets? Probably. It certainly seems that most prediction markets—even real money ones—establish payoffs in advance of any game or contest (i.e., you stand to win the face value of any certificates you hold at the expiry of trading in the associated claim), and those payoffs are not determined by the number of participants or the fees they've paid. That takes care of section 5362(1)(E)(ix)(I).
Whether or not prediction markets satisfy 5362(1)(E)(ix)(II), which was evidently written with fantasy sports teams in mind, poses a closer question. Certainly, the winning outcomes on a prediction market reflect the relative knowledge and skill of the participants. Are they also "determined predominantly by accumulated statistical results of the performance of individuals . . . in multiple real-world . . . events"? In the case of prediction markets in elections, almost certainly. In the case of prediction markets in science claims, possibly. Claims on such markets would, to play it safe, have to relate not to mere abstract phenomena but rather to the performance of particular scientists. Write claims that treat science (or movie making, or anything else) like a spectator sport, in other words, and you will probably satisfy the requirements of UnInGEn-ious Act section 5362(1)(E)(ix)(II).
The last condition, in section 5362(1)(E)(ix)(III), presents no problem for prediction markets.
Even if prediction markets do not qualify for safe harbor in UnInGEn-ious Act section 5362(1)(E)(ix), moreover, it does not follow that the Act renders them illegal. Recall, after all, that the Act reaches only bets or wagers. As I've argued elsewhere, prediction markets generally do not deal in bets or wagers as defined in applicable state or federal laws. Even the UnInGEn-ious Act makes that clear; section 5362(1)(A) defines "bet or wager" as, in relevant part, "the staking or risking by any person of something of value upon the outcome of a contest of others, a sporting event, or a game subject to chance . . . ."
Bottom line: The UnInGEn-ious Act does not expand U.S. law to outlaw prediction markets and offers a safe harbor expressly exempting properly structured prediction markets from the Act's reach.
[Crossposted, in part, at Midas Oracle.]
Tuesday, August 29, 2006
Prediction Markets Whup on Copyrights and Patents
I've just posted to SSRN my forthcoming paper, "Prediction Markets for Promoting the Progress of Sciences and the Useful Arts." It will appear in vol. 14 of the George Mason Law Review. You can download a pretty-to-close-to-final draft at here. The abstract:
Copyrights and patents promote only superficial progress in the sciences and useful arts. Copyright law primarily encourages entertaining works, whereas patent law mainly inspires marginal improvements in mature technologies. Neither form of intellectual property does much to encourage basic research and development. Essential progress suffers.
Prediction markets offer another way to promote the sciences and useful arts. In general, prediction markets support transactions in claims about unresolved questions of fact. A prediction market specifically designed to promote progress in the sciences and useful arts - call it a scientific prediction exchange or SPEx - would support transactions in a variety of prediction certificates, each one of which promises to pay its bearer in the event that an associated claim about science, technology, or public policy comes true. Like other, similar markets in information, a scientific prediction exchange would aggregate, measure, and share the opinions of people paid to find the truth.
Because it would reward accurate answers to factual questions, a SPEx would encourage essential discoveries about the sciences and useful arts. Researchers and developers in those fields could count on the exchange to turn their insights into profit. In contrast to copyrights or patents, therefore, a SPEx would target fundamental progress. Furthermore, and in contrast to copyrights and patents, the exchange would not impose deadweight social costs by legally restricting access to public goods. To the contrary, a scientific prediction exchange would generate a significant positive externality: Claim prices that quantify the current consensus about vital controversies.
This article measures copyright and patent law against the Constitution's call for promotion of the Progress of science and useful Arts, to find those traditional forms of intellectual property lacking. As a cure for that policy failure, it suggests scientific prediction exchanges. Given that such exchanges offer the promise of large net public and private benefits, why don't they already thrive in the United States? Because the laws written for commodity futures, securities, and gambling markets cast a pall of legal uncertainty over scientific prediction exchanges. To ease that unwarranted burden, the article explores a variety of strategies designed to guarantee the legality of scientific prediction exchanges. The article concludes with an all-too-apt illustration of how legal risks can discourage prediction markets from promoting the progress of science and the useful arts.
Wednesday, March 29, 2006
Architzel on Legality of Prediction Markets
The indefatigable Chris F. Masse directed my attention to a new analysis of the legality of prediction markets: Paul Architzel, Event Markets Evolve: Legal Certainty Needed, FUTURES INDUSTRY, March/April 2006, at 50-53. By my count, that marks the fourth author and fourth paper to give serious consideration to the question. (The others: my 1998 paper in CHAPMAN LAW REVIEW, my recent draft paper, and Hahn & Tetlock's forthcoming J. REG. ECON. paper.)
Architzel seems very well qualified to comment on commodity futures regulations. He recently joined the financial services and products practice group of the blue-chip law firm, Alston & Bird. Prior to that gig, he served as chief regulatory counsel of EurexUS for three years and as chief counsel of the CFTC's Division of Market Oversight for over 20 years. His article, though brief and unfootnoted, speaks with authority about commodity futures law.
As his article's title indicates, Architzel thinks that the uncertainty of U.S. law currently hinders the development of prediction markets. He flags three recent developments that hint that the CFTC might consider extending its jurisdiction to cover such markets. (I blogged about one of those regulatory incidents—the settlement between TEN and the CFTC—and concluded that it showed the CFTC recognizes that it does not have regulatory authority over prediction markets.) Architzel goes on to suggest four possible legal outcomes of operating a prediction market under U.S. law. Unsurprisingly, each involves either conformity with CFTC regulations or the risk of "operating in violation of Commission rules." P. 52.
Architzel is a CFTC man, through and through. He says next to nothing about whether gambling regulations reach prediction markets, and nothing about whether securities regulations do. Just as he finds that legal uncertainty about CFTC action hinders prediction markets, Architzel looks to the CFTC for a cure. He writes, "The CFTC could provide greater certainty to event markets in a manner that would also nurture their unique nature by promulgating a non-exclusive safe-harbor rule under which event markets could operate." P. 53.
In assuming that the CFTC could and should have jurisdiction over prediction markets, Architzel's analysis largely mirrors that of Hahn & Tetlock. I critiqued that approach in my draft paper, excerpting the most relevant part in a blog post. In very brief, I both doubt that the CFTC has jurisdiction over prediction markets and that CFTC regulation of them would generate net benefits.
Even sticking with the assumption that the CFTC has jurisdiction over prediction markets, however, Architzel's otherwise very solid analysis of commodity futures regulation raised my eyebrows twice. First when he said, at p. 52, that if the events subject to trading on prediction markets were "considered to be 'commodities,' they might even be offered on designated contract markets without restriction as to market participant." Would the CFTC really give so unusual and broad a definition to "commodity" under the Commodity Exchange Act? I must admit that I'd never even considered that possibility. Second, I was surprised and disappointed that Architzel did not discuss the "hybrid instrument predominantly a security" exclusion in CEA § 2(f)(1), a legal loophole I discuss in my draft paper.
I've written to Architzel on both counts, but as yet heard no reply. Those represent mere quibbles, though. By and large, I've very happy to have another ally in the struggle to clear the legal path for prediction markets.
Wednesday, March 15, 2006
What about "SEx"?
I earlier asked for comments on my draft paper, The Prediction Exchange: Progress in Promoting the Sciences and Useful Arts. Several people pitched in, mostly via private emails. I don't know if that's due to the difficulties of commenting via Blogger's interface or because my correspondents had the decorum to not call me "blithering idiot" publicly. I take comments seriously (apart from those of the "blithering idiot" category; I'm clearly more in the "raving lunatic" line). I here share a couple that I thought especially worthy of others' consideration
Firstly, one of my law professor friends worried that a prediction market would create an unseemly incentive for researchers to hide their discoveries, rather than to promulgate them. He reasoned that a discoverer of new and material information would want to play the market rather than tip her hand. I replied along these lines:
Rather than non-disclosure, I think we would see this pattern: Trade first and then talk. Suppose, for instance, that you have developed a proof of Goldbach's conjecture. See, e.g., the GBch claim currently traded on the Foresight Exchange. You load up on the GBch claim, currently trading at $.42/share, and then release your findings. The price soars and you sell at a profit.
Yeah, you might try a head-fake, first, to drive down the price. But how are you going to fool the market? And, anyhow, once you start buying the price is going to rise, anyhow. So a profit-maximizer will want to get it up sooner rather than later, to cash in. Time value of money and all that.
Admittedly, though, I cooked up that reply off the top of my head. I am not aware of any more complete and careful reply to my correspondent's good question.
Secondly, a few people questioned my use of "prediction exchange" for "prediction market in claims about the sciences and useful arts." One person thought my label too indistinct; another argued that an "exchange" is a collection of markets. I replied, (again paraphrasing):
I agree that "prediction exchange" is not ideal. I've not yet found an alternative term I'm really happy with, though. Tacking "science" or "technology" or "useful art" to "market" or "exchange" strikes me as too narrow. "Sciences and useful arts exchange" would work, but . . . ugh. So I satisfied myself, for the purposes of this paper, with trying to give "prediction exchange" and "prediction market" distinct and clear meanings within the paper's bounds. But, agreed, it would be far better to come up with some label that works more generally.
What about "Science and Technology Exchange?" It's wordy, but I could use an acronym—"STEx" maybe. Or "SciEx." I don't guess "SEx" would be very appropriate, though it would certainly get me a lot of SSRN downloads.
That terminological problem really bugs me. I get pretty worked up about the rhetorical uses of names in any event. Here, it seems both especially important and especially difficult to get the name right. On top of all else, I'm in that post partum stage, after having loosed a writing on the world, where I find myself particularly susceptible to niggling doubts. I wonder, "Did I blither where I should have raved?"
Thursday, March 09, 2006
Money Mouth
I'm wrapping up my most recent paper on prediction markets, "The Prediction Exchange: Progress in Promoting the Sciences and Useful Arts." By way of the grand finale, I'm toying with putting in the conclusion this passage:
I hereby offer to sell for $.10 each up to 1000 prediction certificates, each of which contains the following promise: "The bearer of this certificate is entitled to receive $1.00 from Tom W. Bell if, before January 1, 2010, a federal or state court in the United States holds that a prediction exchange fitting the definition set forth in [this article] violates state or federal gambling laws or falls within the jurisdiction of the SEC or CFTC." The certificates will contain some additional fine print, naturally. That promise captures the heart of my prediction, however. It moreover shows that I am willing to risk $1000 on my claim that, all told, prediction markets should not be illegal under U.S. law. More exactly, my asking price shows that I think it only 10% likely that my prediction errs.
So, dear readers and fellow prediction market geeks, what do you think of that? Yes, I know it's typically nutty; I mean only to ask if this admittedly rough proxy for a real prediction market contains some fatal flaw. Also, you may have some ideas about what I need to put in the fine print. I certainly to discuss, in a footnote, the need to name judges, the choice of law provision, the fact that I can delegate my duty. I may even include a sample certificate. I doubt I can easily go as far as Robin Hanson did, so many years ago, in his Extropy article on prediction markets. He included, in each hardcopy, a real coupon!