Difference between revisions of "Prediction Markets"

From The Internet: Issues at the Frontier (course wiki)
Jump to navigation Jump to search
Line 11: Line 11:
 
'''I. What Prediction Markets Do Now'''
 
'''I. What Prediction Markets Do Now'''
  
The most high-profile examples of prediction marketsthe Iowa Electronic Markets and Intradestarted by focusing primarily on predicting election outcomes and related political and financial events. Now they have expanded to cultural (Oscars) and technological (X Prize) events as well.  The status of the commercial prediction markets is uncertain; for example, [http://en.wikipedia.org/wiki/TradeSports Tradesports] announced recently that it [http://www.tradesports.com/ is closing].  And questions remain about the [http://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID1134563_code183716.pdf?abstractid=1134563&mirid=1 legal status] of prediction markets, whether the CTFC will [http://www.cftc.gov/lawandregulation/federalregister/proposedrules/2008/e8-9981.html regulate] them, and whether they will be [http://findarticles.com/p/articles/mi_6716/is_4_27/ai_n29450615/pg_1?tag=artBody;col1 taxed].    
+
The most high-profile examples of prediction marketsthe Iowa Electronic Markets and Intradestarted by focusing primarily on predicting election outcomes and related political and financial events. Now they have expanded to cultural (Oscars) and technological (X Prize) events as well.  The status of the commercial prediction markets is uncertain; for example, [http://en.wikipedia.org/wiki/TradeSports Tradesports] announced recently that it [http://www.tradesports.com/ is closing].  And questions remain about the [http://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID1134563_code183716.pdf?abstractid=1134563&mirid=1 legal status] of prediction markets, whether the CTFC will [http://www.cftc.gov/lawandregulation/federalregister/proposedrules/2008/e8-9981.html regulate] them, and whether they will be [http://findarticles.com/p/articles/mi_6716/is_4_27/ai_n29450615/pg_1?tag=artBody;col1 taxed].
  
Rather than focusing on the traditional markets, however, we want to focus on future applications of prediction markets, particularly their possible use by government or by government-industry collaboration.  We'd like to explore in particular applications that are likely to be controversial.   
+
Several weeks ago the class started an experiment with its own prediction markets.  Here was the plan: "Everyone will create their own prediction market contract via intrade.net.  Once you set up an account there, you can then create you own contract at http://www.intrade.net/market/create/start.faces.  Your contract can be about anything you like (doesn't have to be legal), but it should conclude before our class date, April 27, 2009.  The idea is (1) for everyone to get a feel for how prediction markets operate; (2) to see what kinds of contracts get enough volume to be successful and what kinds don't; (3) to see how accurate the predictions are, how they change over time, etc.   
  
The focus will be three cases that we think raise interesting legal and ethical questions.  
+
"Once you've signed up and created a contract -- please do so by Friday, March 20 -- you can invite other people in the class to bet on your contract to give it some initial starting volume.  (Intrade.net gives everyone $10k in play money upon signing up, so obviously if people want to bet on the wider world's contracts, that's great too).  Everyone should list the contracts they've created below to facilitate class participation on intrade (and so that we can generate a variety of different kinds of contracts). At the actual class session on April 27, everyone should come in prepared to briefly discuss what happened with their contract (feel free to create more than one)."
  
* Crime rate predictions, a la [http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1118931 this proposal]
+
The experiment was, like Apollo 13, a successful failure.  The markets got no traffic, and the predictions did not move.  There are several explanations for this result, some of which definitely played a role and some of which may have.  First, intrade.net is a fake money site.  The lack of a financial incentive is, of course, a substantial disincentive to would-be bettors.  But this explanation cannot be sufficient, because the whole intrade.net site--not just our class-constructed markets--is based on the same fake money system.  In fact, what is remarkable is how active and accurate some of the fake money markets are.  Second, though, our markets were placed in the user-created market ghetto of the intrade.net site. Traffic there is generally lower than on the rest of the site, and activity is correspondingly reduced. Could this be an example of a coordination problem? Third, some of the topics chosen were obscure; an efficient market would need enough investors interested in the subject to get trading volume sufficiently high.  Fourth, and probably most devastating, the betting mechanism only allows an up-down vote.  This limitation takes away one of the essential comparative benefits of prediction markets over voting: it does not account for intensity of preference.  The bettor who thinks the going rate is 1% off sends the same signal as the bettor who thinks the rate is 50% off.  A well-designed prediction market would never be designed this way.
  
* Google's flu-tracking application (where, as Professor Zittrain noted, the predictors aren't even aware that their knowledge is being harvested)
+
Required Reading for Part I
 
+
* [http://bpp.wharton.upenn.edu/jwolfers/Papers/Predictionmarkets.pdf <i>Prediction Markets</i>]
* the failed DARPA [http://www.cnn.com/2003/ALLPOLITICS/07/29/terror.market/index.html terrorist futures] market
+
Justin Wolfers and Eric Zitzewitz | 18 Journal of Economic Perspectives 107 (2004) | (19 pages, essential background--only one equation for the math-phobic)
 
 
'''Guests'''
 
 
 
Justin Wolfers, Economist (confirmed)
 
 
 
Hal Varian, Google (possibly, on videoconference)
 
 
 
'''Concrete Questions'''
 
 
 
* Which of these applications are most likely and desirable?
 
 
 
* Should the government be involved in administering prediction markets at all? Should it regulate them?
 
 
 
* What ethical concerns do we have about prediction markets of the future, and how might we address them? Can design of the markets help mitigate concerns? Are some more fundamental? 
 
 
 
'''Tech Tweak/Experiment'''
 
 
 
Everyone will create their own prediction market contract via intrade.net.  Once you set up an account there, you can then create you own contract at http://www.intrade.net/market/create/start.faces.  Your contract can be about anything you like (doesn't have to be legal), but it should conclude before our class date, April 27, 2009.  The idea is (1) for everyone to get a feel for how prediction markets operate; (2) to see what kinds of contracts get enough volume to be successful and what kinds don't; (3) to see how accurate the predictions are, how they change over time, etc. 
 
  
Once you've signed up and created a contract -- please do so by Friday, March 20 -- you can invite other people in the class to bet on your contract to give it some initial starting volume.  (Intrade.net gives everyone $10k in play money upon signing up, so obviously if people want to bet on the wider world's contracts, that's great too).  Everyone should list the contracts they've created below to facilitate class participation on intrade (and so that we can generate a variety of different kinds of contracts). At the actual class session on April 27, everyone should come in prepared to briefly discuss what happened with their contract (feel free to create more than one). 
 
  
'''Readings'''
+
'''II. Novel Applications of Prediction Markets'''
  
* [http://bpp.wharton.upenn.edu/jwolfers/Papers/Predictionmarkets.pdf <i>Prediction Markets</i>]
 
Justin Wolfers and Eric Zitzewitz | 18 Journal of Economic Perspectives 107 (2004) | (19 pages, essential background--only one equation for the math-phobic) 
 
  
 +
Required Reading for Part II
 
* [http://www.nytimes.com/2008/11/12/technology/internet/12flu.html <i>Google Uses Searches to Track Flu’s Spread</i>]
 
* [http://www.nytimes.com/2008/11/12/technology/internet/12flu.html <i>Google Uses Searches to Track Flu’s Spread</i>]
 
Miguel Helft | New York Times | November 11, 2008 | (1 page, useful diagram)
 
Miguel Helft | New York Times | November 11, 2008 | (1 page, useful diagram)
Line 57: Line 37:
 
Carl Hulse of the NYT and Wire Services | July 2003 | (~6 pages)
 
Carl Hulse of the NYT and Wire Services | July 2003 | (~6 pages)
  
 +
 +
'''III. Skeptical Concerns'''
 +
 +
 +
Required Reading for Part III
 
* [http://ftp.fas.org/sgp/congress/2003/s072803.html Dorgan Speech] [http://www.fas.org/sgp/congress/2003/s072903.html Daschle Speech]
 
* [http://ftp.fas.org/sgp/congress/2003/s072803.html Dorgan Speech] [http://www.fas.org/sgp/congress/2003/s072903.html Daschle Speech]
 
Discussion of the DARPA Markets in the Senate | July 28 and 29, 2003 | (2 short comments, mostly for humor value)   
 
Discussion of the DARPA Markets in the Senate | July 28 and 29, 2003 | (2 short comments, mostly for humor value)   
Line 64: Line 49:
  
  
 
+
'''Optional Reading'''
'''Optional Readings'''
 
  
 
**A brief paper describing the failed DARPA terrorism futures market, by a Professor at the Naval Postgraduate School. [http://www.ccc.nps.navy.mil/si/sept03/terrorism.pdf Robert Looney, <i>DARPA’s Policy Analysis Market for Intelligence: Outside the Box or Off the Wall?</i>, STRATEGIC INSIGHTS (2003).]  
 
**A brief paper describing the failed DARPA terrorism futures market, by a Professor at the Naval Postgraduate School. [http://www.ccc.nps.navy.mil/si/sept03/terrorism.pdf Robert Looney, <i>DARPA’s Policy Analysis Market for Intelligence: Outside the Box or Off the Wall?</i>, STRATEGIC INSIGHTS (2003).]  

Revision as of 08:06, 20 April 2009

Topic Owners: Matthew, Elisabeth

back to syllabus

Precis

Our seminar this week aims to discuss novel and/or controversial applications of prediction markets. Our study will be organized into three parts. First, before class everyone will get acquainted with prediction markets as they now exist and understand the basic economic theory behind their success--this page and the first required reading (the Wolfers/Zitzewitz article) will cover that background. We also tried a (failed!) experiment with creating our own prediction markets on intrade.net that inadvertently taught a lesson about how difficult it is to get efficient information markets (more below). Second, we will examine a few proposals for how prediction markets might be used at the frontier. The readings cover Professor Wolfers' crime prediction market, the DARPA project, and the Google Flu Tracker. In class, Professor Wolfers' will expand on the crime prediction market, and Bo Cowgill of Google will add some insight into Google's internal prediction markets. Third and finally, we will try to generalize from these examples some basic principles about what prediction markets can and cannot do well. At this point, we will consider skeptical arguments, both the set of (weak but popular) moral objections and more sophisticated practical/theoretical worries. Ultimately, the class will come to some conclusions about what types of uses prediction markets might have, perhaps inventing some ideas of our own.

I. What Prediction Markets Do Now

The most high-profile examples of prediction marketsthe Iowa Electronic Markets and Intradestarted by focusing primarily on predicting election outcomes and related political and financial events. Now they have expanded to cultural (Oscars) and technological (X Prize) events as well. The status of the commercial prediction markets is uncertain; for example, Tradesports announced recently that it is closing. And questions remain about the legal status of prediction markets, whether the CTFC will regulate them, and whether they will be taxed.

Several weeks ago the class started an experiment with its own prediction markets. Here was the plan: "Everyone will create their own prediction market contract via intrade.net. Once you set up an account there, you can then create you own contract at http://www.intrade.net/market/create/start.faces. Your contract can be about anything you like (doesn't have to be legal), but it should conclude before our class date, April 27, 2009. The idea is (1) for everyone to get a feel for how prediction markets operate; (2) to see what kinds of contracts get enough volume to be successful and what kinds don't; (3) to see how accurate the predictions are, how they change over time, etc.

"Once you've signed up and created a contract -- please do so by Friday, March 20 -- you can invite other people in the class to bet on your contract to give it some initial starting volume. (Intrade.net gives everyone $10k in play money upon signing up, so obviously if people want to bet on the wider world's contracts, that's great too). Everyone should list the contracts they've created below to facilitate class participation on intrade (and so that we can generate a variety of different kinds of contracts). At the actual class session on April 27, everyone should come in prepared to briefly discuss what happened with their contract (feel free to create more than one)."

The experiment was, like Apollo 13, a successful failure. The markets got no traffic, and the predictions did not move. There are several explanations for this result, some of which definitely played a role and some of which may have. First, intrade.net is a fake money site. The lack of a financial incentive is, of course, a substantial disincentive to would-be bettors. But this explanation cannot be sufficient, because the whole intrade.net site--not just our class-constructed markets--is based on the same fake money system. In fact, what is remarkable is how active and accurate some of the fake money markets are. Second, though, our markets were placed in the user-created market ghetto of the intrade.net site. Traffic there is generally lower than on the rest of the site, and activity is correspondingly reduced. Could this be an example of a coordination problem? Third, some of the topics chosen were obscure; an efficient market would need enough investors interested in the subject to get trading volume sufficiently high. Fourth, and probably most devastating, the betting mechanism only allows an up-down vote. This limitation takes away one of the essential comparative benefits of prediction markets over voting: it does not account for intensity of preference. The bettor who thinks the going rate is 1% off sends the same signal as the bettor who thinks the rate is 50% off. A well-designed prediction market would never be designed this way.

Required Reading for Part I

Justin Wolfers and Eric Zitzewitz | 18 Journal of Economic Perspectives 107 (2004) | (19 pages, essential background--only one equation for the math-phobic)


II. Novel Applications of Prediction Markets


Required Reading for Part II

Miguel Helft | New York Times | November 11, 2008 | (1 page, useful diagram)

M. Todd Henderson, Justin Wolfers, and Eric Zitzewitz | Olin Working Paper No. 402 | April 2008 | (5 pages, just read abstract and introduction)

Carl Hulse of the NYT and Wire Services | July 2003 | (~6 pages)


III. Skeptical Concerns


Required Reading for Part III

Discussion of the DARPA Markets in the Senate | July 28 and 29, 2003 | (2 short comments, mostly for humor value)

Student Note | 122 Harvard Law Review 1217 (2009) | (21 pages, actually good critical arguments)


Optional Reading

  • Professor Sunstein's Infotopia
  • Cass R. Sunstein, Group Judgments: Statistical Means, Deliberation, and Information Markets, 80 N.Y.U. L. Rev. 962 (2005)

Class-Constructed Markets

Andrew: Will General Motors file for Chapter 11 by April 20, 2009? http://www.intrade.net/market/detail/?contractId=338028#GM_Files_for_Chapter_11_by_4/20/09%3F

  • Gwen: I'm having trouble getting an account set up. Do you have to mail in a check? I get this error message when I try to use credit cards: "Unfortunately your bank has refused to authorise your deposit. Most US-based members are no longer able to fund their accounts by credit card due to the passage of recent legislation through the US Congress. Many non-US banks have also adopted a similar policy and will no longer authorise credit card transactions with certain classes of merchants. For more information about your attempted deposit please contact your bank. If you require assistance with funding your account by another method please e-mail us at help@intrade.com . We are happy to accept deposits by check, money order or bank transfer. Please see your My Account screen for more details." What law is this, and how recently did it change? [Elisabeth: Intrade.net is a free site -- you shouldn't have to use your credit card! You might be at the wrong site?]