Wireless broadband technology is now a vital part of our daily lives and plays a crucial role in industrial competitiveness. Thus, ensuring an efficient relocation of spectrums to meet the exploding demands will be an important public policy agenda for years to come. To achieve this objective, there have been significant innovations in market design such as simultaneous ascending auctions, package auctions, and incentive auctions. Nevertheless, the information about the public policy issue and market design algorithms are not yet widely available and the lack of information creates barriers in education, research, public awareness, and policy formation. The PI and the development team will leverage their multi-disciplinary and cross-domain expertise in economics and computer science to address these challenges.

The goals of this project are to provide: 1) a software that implements prototypes of major spectrum allocation algorithms, 2) detailed information about their definition, theoretical properties, empirical application, and experimental results, 3) tutorials and lesson plans for the spectrum allocation policy and the role of public policy, 4) case studies of spectrum allocation policies in the US, Europe, and Asia describing the application of economic theory, data analysis, and policy decisions, and 5) experimental and theoretical studies of the spectrum allocation mechanism employing the software.

In the short term, this project will enable educators to develop courses for the spectrum allocation policy and market designs, students to actually work with these mechanisms, researchers in labs and in the field to conduct experiments and simulate auction outcomes, and policy makers to understand current best practices in the spectrum policy. In the long term, the outcomes of this research will improve public awareness of the spectrum policy.

Project Report

**The backgroud of the project** Wireless broadband technology is now a vital part of our daily lives and plays a crucial role in industrial competitiveness. Thus, ensuring an efficient relocation of spectrums to meet the exploding demands will be an important public policy agenda for years to come. To achieve this objective, there have been significant innovations in market design such as simultaneous ascending auctions, package auctions, and incentive auctions. Nevertheless, the information about the public policy issue and market design algorithms are not yet widely available and the lack of information creates barriers in education, research, public awareness, and policy formation. The PI and the development team leverage their multi-disciplinary and cross-domain expertise in economics and computer science to address these challenges. **The outcomes of this project** 1. The PI developed softwares that implement prototypes of major spectrum allocation algorithms and also databases on Japanese and US markets. Spectrumwiki at http://spectrumpolicy.kazumori.net Ttree at http://ttree.kazumori.net Tokyo Research Data Services at http://trds.kazumori.net 2. The PI has developed 4 research papers on the issue of incentive auctions and spectrum auctions Kazumori, E. (2013a). "Simultaneous Heuristic Auctions," Working Paper. Kazumori, E. (2013b). "Experimental Studies of Clock-Proxy Auctions," Working Paper. Kazumori, E., and Y. Belch (2013c). "On the Design of Reverse Auctions with Multiple Bid Options for the US Incentive Auctions: Generalized Heuristic Threshold Auctions," Working Paper. Kazumori, E., and Y. Belch (2013d). "Descending Clocks, Ascending Clocks, and Closing Rule: An Open-Source Prototype Software for Incentive Auctions," Working Paper. 3. The PI has developed 6 papers related to the project. Kazumori, E. (2013b): Deleveraging, Liquidity Risk, and Asset Prices: Evidences from US and Japanese Stock Markets, 1977-2009. Kazumori, E. (2013c): "What Drives Stock Returns of Apple, Google, Microsoft and Oracle? Industry Structure and Stock Returns in the US Computer Industry, 1965-2012.," Working Paper. Kazumori, E. (2013d): "Double Auction Markets with Interdependent Values: An Asymptotic Equivalence Approach," Working Paper. Kazumori, E. (2013e): "Optimal Auction Design for Heterogeneous Object," Working Paper. Kazumori, E. and Tchuindjo, L. (2013a): "Primary Dealers, Indirect Bidders, and Direct Bidding: A Structural Model of US Treasury Auctions," In Preparation. Kazumori, E. and Tchuindjo, L. (2013b): "Treasury Inflation Protected Securities: The Role of Margin Requirements," In Preparation. 4. The PI has made 9 referred presentations: 2013 North American Winter Meeting of the Econometric Society, San Diego (Presentation, Discusstant) The University of Tokyo: Empirical Microeconomics Workshop (Presentation) 2013 Second Cambridge Area Economics and Computation Day, MIT. (Presentation) 5th Israel Game Theory Conference, Tel-Aviv. (Presentation) 2013 North American Summer Meeting of the Econometric Society, USC (Presentation) 2013 Stony Brook Game Theory Conference (Presentation) 2013 Midwest Theory Conference, Michigan (Presentation) 2013 NSF EARS Workshop, Washington DC (Presentation) 2014 North American Winter Meeting of Econometric Society, Pennsylvania (Planned, Presentation, Discusstant) 5. The PI has developed an educational software that incorporates technology, administrative process, economics, and public policy. The content consists of Spectrum Regulatory Agencies Administriative Process 1994 Nationwide PCS auctions Economics of Spectrum Auctiond Further Global Development Incentive Auctions Economics of Incentive Auctions Experiencing Incentive Auctions Toward a Spectrum Future ** Intellectual Merits** 1. The incentive auctions software is the first implementation of the incentive auctions process defined in FCC Incentive Auctions NPRM. It demonstrates the feasibility and practicality of the auction design. 2.The theoretical results of heuristic auctions with multiple relinquishment options demonstrate robustness and universality of Milgrom-Segal heuristic auctions design. 3. The experimental results on package auctions validate the previous theoretical predictions and also show the practical importance of communication costs and subject characteristics. 4. The empirical results on liquidity risks in the Japanese and US markets validate previous results on imporatance of liquidity in asset pricing and also sheds lights on the comparison between Japanese and US markets. 5. The empirical and theoretical results on stock prices in computer industries connect the previous IO literature and asset pricing literature and also extend the consumption based asset pricing model to allow imperfect competition among companies. 6. The theoretical rsults on double auctions generalizes previous results on double auctions with interdependent values and also integrate one-sided auctions and double auctions 7. The theoretical results on optimal auctions generalizes previous results on optimal auctions to allow multidimensional signals and heterogenous objects and also integrates optimal auctions literatures and nonlinear pricing literature. ** Broader Impacts** 1. The experiment software complements Ztree softwares and allows researchers to conduct experiments and teaching on auctions and market design. 2. The Spectrumwiki software integrates technology, public policy, and economics and can serve education and tutorial purposes. 3. The TRDS software has more than 120,000 accesses and also offers from Bloomberg for globral distribution of data.

Agency
National Science Foundation (NSF)
Institute
Division of Social and Economic Sciences (SES)
Type
Standard Grant (Standard)
Application #
1247988
Program Officer
Michael Reksulak
Project Start
Project End
Budget Start
2012-09-15
Budget End
2013-08-31
Support Year
Fiscal Year
2012
Total Cost
$65,000
Indirect Cost
Name
Suny at Buffalo
Department
Type
DUNS #
City
Buffalo
State
NY
Country
United States
Zip Code
14260