The dark side of the internet, in a nutshell?

As incentives to take higher actions increase—due to higher stakes or more manipulable signaling technology—more information is revealed about gaming ability, and less about natural actions. We explore a new externality: showing agents’ actions to additional observers can worsen information for existing observers. Applications to credit scoring, school testing, and web searching are discussed.

That is from a forthcoming JPE paper “Muddled Information,” by Alex Frankel and Navin Kartik.


Information overload? Too much of a good thing? Reminds me of the study that showed excessive financial trading can be shown to be detrimental to a countries GDP after a certain point. Extra liquidity is outweighed by too much information seeking.

The paper has nothing to do with the internet. It is 100% about agents taking actions to game the signaling in their environment therefore making old signals less reliable for all participants. A signal is supposed to give you information about what they call the "natural action" of an agent, meaning what they will do if they believe they are not being observed. When the signals are well known, then these actions are gamed and actions taken differ from their natural action. That is why from time to time Google changes its algorithms, the SAT changes its criteria, and FICO changes their scoring to cite the three major examples in the paper. It is another restatement of the ideas expressed in Campbell's Law, Goodhart's Law, or the Lucas Critique but I did not see any of those ideas mentioned.

Save yourself $20. A copy of Frankel's paper is on his faculty site free of charge:

lol@your first sentence, followed by the other ones.

we think outside of the stem fields
farting muddled information is the
predominant postmodern gaming/signaling strategy

This article is very informative through it we can learn many things. Thanks for sharing this information.

Comments for this post are closed