Secret Deliberations

53 Pages Posted: 19 Sep 2015

See all articles by Parimal K. Bag

Parimal K. Bag

National University of Singapore (NUS)

Tridib Sharma

Instituto Tecnológico Autónomo de México (ITAM) - Centro de Investigacion Economica

Date Written: September 18, 2015

Abstract

Sequential advice to a decision maker by two experts, if kept secret from the public, outperforms transparent deliberations. Under secrecy, the prior bias together with the bias due to the lead expert's opinion may or may not generate a conformity bias effect. These effects determine if and when the second expert's recommendation alters the decision maker's choice for the better. The biases may reinforce or oppose each other, leading to either full signal revelation or a beneficial (partial) herding by the second expert. For on average high ability experts, the only way a second expert can be effective is through secret deliberations. The analysis shows complex strategic interactions in sequential deliberations, advancing earlier findings on career concerns and cheap talk advice.

Keywords: Expert advice, career concern, cheap talk, market reputation, deliberation, transparency, secrecy, prior bias, lead opinion bias, conformity bias, signal revelation, beneficial herding, type revelation

JEL Classification: D82, D83, D23, C72

Suggested Citation

Bag, Parimal K. and Sharma, Tridib, Secret Deliberations (September 18, 2015). Available at SSRN: https://ssrn.com/abstract=2662370 or http://dx.doi.org/10.2139/ssrn.2662370

Parimal K. Bag (Contact Author)

National University of Singapore (NUS) ( email )

1E Kent Ridge Road
NUHS Tower Block Level 7
Singapore, 119228
Singapore

Tridib Sharma

Instituto Tecnológico Autónomo de México (ITAM) - Centro de Investigacion Economica ( email )

Av. Camino a Santa Teresa #930
Col. Heroes de Padierna
Mexico City, D.F. 10370
Mexico D.F.
+525 6284197 (Phone)
+525 6284058 (Fax)

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
44
Abstract Views
424
PlumX Metrics