Changes

Jump to navigation Jump to search
no edit summary
These chapters reference:
*Crawford, Vincent P. and Joel Sobel (1982), "Strategic Information Transmission", Econometrica, Vol. 50, No. 6 (Nov.), pp. 1431-1451 [http://www.edegan.com/pdfs/Crawford%20Sobel%20(1982)%20-%20Strategic%20Information%20Transmission.pdf pdf]
*Krishna, Vijay and John Morgan (2001), "A Model of Expertise", The Quarterly Journal of Economics, Vol. 116, No. 2 (May), pp. 747-775 [http://www.edegan.com/pdfs/Krishna%20Morgan%20(2001)%20-%20A%20Model%20of%20Expertise.pdf pdf]
==Abstract==
:<math>2n(n-1)\delta < \theta_{max} - \theta_{min}\,</math>
 
 
There are some things to note:
#Partitions get bigger to the right
#There always exists an equilibrium with <math>n=1\,</math> - it is the babbling equilibrium
#As <math>\delta\,</math> gets smaller, more partitions can be sustained.
#If <math>n\,</math> partitions can be sustained then so can <math>k<n\,</math>.
 
===Welfare considerations===
 
More partitions are more informative and yield greater welfare. The equilibrium with the greatest number of partitions is the one preferred by both the SIG and the policymaker.
 
===Two Lobbies - Like bias===
 
Suppose that <math>0 < \delta_1 \le \delta_2\,</math> so that SIG 1 is more moderate. There are three cases to consider: secret meetings, private meetings and public meetings. With secret meetings the SIGs do not know if the other SIG has met with the policy maker, and with private meetings they know the meeting has taken place but do not know the contents.
 
====Secret Meetings====
 
The SIGs behave as if they alone are providing information and the policy maker combines informations. The combined information is more informative, but there is no actual equilibrium (because of the implicit assumption of no strategic response).
 
====Private Meetings====
 
Using each SIG to discipline the other, the policy maker can gain full relevation. However, this is a fragile equilibrium.
 
====Public Meetings====
 
When information is reported sequentially there can not be full relevation. There can be a partition equilibrium from combining the two reports.
Anonymous user

Navigation menu