Information technology with its rapid growth and ubiquitous nature has come to have a significant impact on individuals, firms and, indeed, on all aspects of society. The technology, and its associated economics, raises a series of interesting questions at the interface of information technology, policy and management. Addressing those issues is of interest to me. Specifically, I have been pursuing problems in the following two research streams (only completed articles are listed):

• Auctions in Electronic Contexts:
  
  

  • Arora, A., Greenwald, A., Kannan, K., and Krishnan R., (2007) "Effects of Information Revelation Policies Under Market Structure Uncertainties," Management Science, 53(8) (Link): use game-theory to analyze information policies when homogeneous bidders are uncertain about the number of competitors. We show that, depending on the policy, bidders alter their behaviors so as to exploit the information revealed and overcome uncertainty. These behaviors lead to a non-monotonic variation of buyer surplus with information revealed. We also demonstrate the importance of considering the nature of uncertainty to assess the impact of an information policy.
    
    

  • Kannan, K. (2009) "Effects of Revelation Policies under Cost Uncertainty," under second round review, Information Systems Research (Link): This study uses game-theory to analyze information policies but in a setting where bidder cost types are heterogeneous and the cost information is private. The analysis identifies two key learning-related effects, which a buyer should consider when choosing a policy: (i) the extraction effect, where a bidder alters his bids in order to learn about his opponent(s); and (ii) the deception effect, where the bids are aimed at preventing the opponent(s) from learning. Either effect can lead to a higher expected price paid by the buyer. Also, because of the learning effects, the social welfare can decrease even if the number of competitors increases.
    
    

  • Greenwald, A., Kannan, K., and Krishnan, R. (2007) "“On Evaluating Information Revelation Policies in e-marketplaces: A Markov Decision Approach," Information Systems Research 20(4), Forthcoming (Link): This paper integrates ideas from distributed-agents, machine-learning, and economics literatures to develop a simulation test-bed that enables comparisons of information policies. The test-bed features suppliers, whose bidding behavior is modeled as a partially observable Markov decision process and learned using a machine-learning technique. The paper also demonstrates the usefulness of the test-bed by comparing information policies in a setting where game-theoretic results are known to be analytically intractable.
    
    

  • Cason, T., Kannan, K., and Siebert. (2009) "An Experimental Study of Information Revelation Policies in Sequential Auctions," working paper: In the laboratory study, we analyze how well real-life subjects, acting as bidders, exhibit the learning effects identified in Kannan (2009). We observe that the subjects behave as if they appreciate the effects. We also observe deviations, which appear consistent with risk aversion and bounded rationality.
    
    

  • Tawarmalani, M., Kannan, K., and De, P. (2009) "Allocating Objects in a Network of Caches: Centralized and Decentralized Analyses" Management Science 55(1) (Link): One of the key contributions of this study is the design of an auction that achieves optimal welfare in a new-technology enabled setting of peer-to-peer caches. The analyses focuses on both centralized and decentralized scenarios when peers share resources. We show that the decentralized peers may be made to coordinate to achieve the optimal allocation outcome through a specific type of an auction.
    
    

  • Balachander S., Kannan, K. and Schwartz, D. (2009) "A Theoretical and Empirical Analysis of Alternate Auction Policies for Search Advertisements," Forthcoming, Review of Marketing Science (Link): Search engine companies such as Mama.com, Yahoo!, and Google differ in terms of the information they use to pick the auction winner(s). We find a counterintuitive result in that the mechanism, which employs more information about the bidders to determine the winner(s), is not always profit maximizing for the search engine. We also provide empirical support for the key assumptions of the theoretical model.
    
    

  On two occasions, in WITS 2000 (for an initial version of Greenwald, et al., 2009) and WITS 2005 (for a preliminary version of Tawarmalani et al., 2009), my papers from this stream have won the best paper awards. AT&T has also expressed interest in extending our ideas on peer-to-peer caches to distribute content on their networks, and has awarded us a research grant.
  
  
• Economics of Information Security:
  
  

  • Kannan, K., and Telang R. (2005) "Markets for Vulnerabilities? Think Again," Management Science, 51 (5) (Link): The paper analyzes a market where vulnerabilities are bought and sold. It shows that without any regulation the market for vulnerabilities generate lower social welfare than a market without even a monopolist. We also show how a social planning organization like CERT (computer emergency response team) can improve welfare by providing vulnerability identifiers with monetary as well as non-monetary benefits.
    
    

  • Zhang, G., Kannan, K., and Tan, Y. (2009) "Optimal Patching Effort Against Internet Worms," working paper: In this paper, we model the worm attack as a stochastic problem and discuss the optimal effort level to be exerted in order to thwart the attack.
    
    

  • Kannan, K., Rees, J., and Sridhar, S. (2007) "Market Reactions to Information Security Breach Announcements" International Journal of Electronic Commerce, 12 (1) (Link): The paper conducts an event study to analyze the effect of security breach media announcements on the market valuations of the breach firms. We find that, at the aggregate level, breach announcements do not have a negative impact on valuations. The study, however, finds a significant negative reaction to the announcements regarding breaches that occurred immediately after September 11, 2001. It also finds differences in investor reactions to events during the dot-com era and after it.
    
    

  • Wang, T., Rees, J., and Kannan, K. (2009) "The Impact of Information Security Disclosures on Market Reactions to Security Breaches": In this recent paper, we study the association between the textual content of the disclosures in public filings, such as 10-K forms, and the breach announcements by integrating ideas from the machine-learning and accounting literatures. Our analysis finds that the textual nature of the disclosures filed by the firms are indeed indicative of future security breach announcements.
    
    

  My research in this stream is getting recognition in the policy and industry circles as well. Rees, Spafford and I were approached by McAfee Inc. to conduct an industry study to understand how perceptions regarding information security vary among senior IT managers across different parts of the world. The study gained attention in the popular press. I was quoted in Forbes; and a White House report cites the study. I am working with Jackie Rees and Chris Forman of Georgia Tech. to explore writing an academic paper on this topic.