Tuesday, April 08, 2008

Course on Demand Revelation at UVa

Here is another course near and dear to my heart. I worked with the developer of the course (Paras Sharma) on immigration reform during the summer of '07. Paras became interested in demand revelation and he developed the course along with his students during the Spring of '08.

Tuesday, October 23, 2007

Letter Concerning Originality in Mechanism Design Theory

From : Nicolaus Tideman Sent : Sunday, October 21, 2007 11:52 PM To : info@kva.se CC : edward clarke Subject : Originality in Mechanism Design Theory To the authors of the Prize Committee's review of Mechanism Design Theory: It is helpful that the Prize Committee provided such an extensive review of mechanism design theory. And yet it is disappointing that the review repeats a common mistake with respect to attribution of originality in one of the important ideas in mechanism design theory. You say (p. 7) "It thus came as a surprise when Edward Clarke (1971) and Theodore Groves (1973) showed that, if there are no income effects on the demand for public goods (technically, if the utility functions are quasi linear), then there exists a class of mechanisms in which (a) truthful revelation of one's willingness to pay is a dominant strategy, and (b) the equilibrium level of the public good maximizes the social surplus. [fn] Footnote: The basic intuition behind the Clarke-Groves mechanism was already present in Vickrey (1961). . . .” The idea that it is possible to motivate people to report accurately the value that they would place on the provision of a public good was understood first by Edward Clarke and was published by him long before anyone else. As you say, Clarke first published his idea in Public Choice in 1971. He published a second use of the idea in Public Prices for Public Products (S. Mushkin, ed.) in 1972. I believe that you will search in vain in Groves (1973) or in Groves's 1970 Berkeley dissertation for any indication that he had any inkling of the possibility of motivating people to report truthfully their preferences for public goods. His dissertation and his Econometrica paper were concerned strictly with incentives in teams. I believe that Groves did become aware, independently of Clarke, of the possibility of using a variation of his idea to motivate the reporting of truthful preferences about public goods. If I remember correctly, Groves told me that the possibility of adapting his idea to public goods came to his attention when a student of his, Martin Loeb, pointed out the possibility, and the two of them then coauthored a paper (1975) explaining the idea. It can be debated whether an independent discovery published four years after an initial publication deserves joint credit for an idea. But if you wish to extend such credit to Groves you should cite not his 1973 paper, but rather the 1975 paper, and extend credit to Loeb as well as Groves. As for Vickrey's contribution, it is true that his 1961 paper shows that it is possible to motivate participants in an auction to report their valuations of an auctioned item truthfully, and this can be understood as an example of the same principle that Clarke employed for public goods and Groves employed for incentives in teams. However, I am not aware of anyone using Vickrey's work to come to and understanding of other possible applications of the idea. Rather, as I understand it, Vickrey’s work was seen as an earlier example of the principle once its general possibilities were understood. Another independent application of the same general idea was Richard Zeckhauser’s suggestion in his 1968 Harvard dissertation that efficient incentives with respect to accidents requires that all parties to an accident be charged the full cost of the accident. This idea can also be seen to be present, obliquely, in Coase’s 1960 paper. The general principle linking all of these ideas is that the generation of efficient incentives requires that people be charged the full marginal social costs (or receive the full marginal social benefits) of their actions, whether these actions are bids in an auction, efforts to make a team more successful, actions that might result in accidents, or the reporting of valuations for a public good. When stated this way, the idea is so fundamental to economics that it is surprising that we were all so surprised by the novel applications of the idea. But we were. And this is the idea that is fundamental to mechanism design. In my view, the expression of this fundamental idea in the context of motivating people to report their preferences for public goods truthfully, as demonstrated first by Clarke, was an important breakthrough that played a key role in helping economists to understand the general principles of mechanism design. Sincerely, Nicolaus Tideman Professor of Economics Virginia Tech References: Clarke, E.H. (1971). “Multipart pricing of public goods,” Public Choice 11, pp. 17-33. Clarke, E.H. (1972). “Multipart pricing of public goods: An example,” pp. 125-30 in S. Mushkin (ed.), Public Prices for Public Products, The Urban Institute, Washington, 1972. Coase R.H. (1960). “The Problem of Social Cost,” Journal of Law and Economics 3, pp. 1-44. Groves T. (1970). “The Allocation of Resources under Uncertainty,” dissertation, University of California at Berkeley. Groves, T. (1973). “Incentives in Teams,” Econometrica 41, pp. 617-31. Groves, T. and Loeb, M. (1975). “Incentives and Public Inputs,” Journal of Public Economics 4, pp. 211-26. Vickrey, W. (1961). “Counterspeculation, Auctions, and Competitive Sealed Tenders,” The Journal of Finance 16, pp. 8-37. Zeckhauser R. (1968). “Studies in Interdependence,” dissertation, Harvard University.

Monday, January 08, 2007

I started this blog because it will probably receive more visibility than one currently at Xanga.com I use the blog mainly to discuss applications of the "demand revealing process" and Vickrey-Clarke-Groves (VCG) mechanisms. The latter are of particular interest in online advertising discussed on my Xanga site. See Link Below. My main website is Public Goods, Good Government (see link below). Links: Ed Clarke's Xanga Blog Public Goods, Good Government The following site has a collection of interesting materials on how the existing online advertising system works and the issues involving "truthful revelation" through VCG mechanisms. Auctions and Inefficiencies in Online Advertising

Tuesday, January 02, 2007

Course on Sponsored Search at UPenn

This appears to be a wonderful course. Has links to key articles and tries to blend science and practice in this rapidly developing field. In the process, VCG mechanisms are getting a lot of attention. "Keyword auctions for sponsored search, such as those offered by Google and Yahoo!, have become profitable and influential. They are also a fascinating new kind of market --- namely, a market in natural language phrases. While the structure of these new markets is evolving rapidly, there is also a diverse and nascent science emerging around the challenges and issues they raise. In this seminar course we will attempt to survey the many academic and commercial views of sponsored search that are under rapid development. The list of candidate topics includes an examination of the details of the various commercial auction mechanisms, their pricing schemes, and the analytic tools and data avaiable to advertisers to track and optimize their keyword portfolios; game-theoretic analyses of the equilibrium and other properties of keyword auctions; and interactions between the economic/financial and natural language aspects of sponsored search."