This paper clarifies why editors of academic journals should share with the
ir referees the information about the number of referees they consult and t
he decision rule they apply. Our analysis also rationalizes the common ques
tionable phenomenon of editors who seem to distort the yes or no recommenda
tions of their referees. The editors request a recommendation of whether to
accept or reject the paper as well as an assessment of the paper. The edit
ors need the complete reports to make the appropriate correction of the ref
erees' final recommendations.