Review of Accounting Studies, 6, 77–108, 2001
2001 Kluwer Academic Publishers. Manufactured in The Netherlands.
Renegotiation and Relative Performance Evaluation:
Why an Informative Signal May Be Useless
ANDREW TZELUNG YIM
Department of Accounting,SchoolofBusinessandManagement,Hong Kong University of Science and Technology,
Clear Water Bay, Hong Kong
Abstract. Although Holmstr¨om’s informativeness criterion provides a theoretical foundation for the controllabil-
ity principle and interﬁrm relative performance evaluation, empirical and ﬁeld studies provide only weak evidence
on such practices. This paper reﬁnes the traditional informativeness criterion by abandoning the conventional
full-commitment assumption. With the possibility of renegotiation, a signal’s usefulness in incentive contracting
depends on its information quality, not simply on whether the signal is informative. This paper derives conditions
for determining when a signal is useless and when it is useful. In particular, these conditions will be met when the
signal’s information quality is either sufﬁciently poor or sufﬁciently rich. (JEL C72, D82).
Keywords: informativeness, monitoring, renegotiation, principal-agent model
Holmstr¨om’s (1979) informativeness criterion is a seminal result in agency theory.
tells us that a signal is useful in incentive contracting if and only if it is informative,
i.e., conveys information not already included in the outcome of the action. This criterion
has been applied to understand management control issues including the controllability
principle and relative performance evaluation. Field studies, such as those by Merchant
(1987) and Maher (1987), however, have not conﬁrmed the consistent application of the
informativeness criterion. Merchant found that while the informativeness criterion suggests
the controllability principle, i.e., evaluation of a manager’s performance should be based
only on factorscontrollable by the manager, the principle is often disregardedin practice. As
an application of the principle, interﬁrm relative performance evaluation (RPE) should be
used to ﬁlter the common uncertainty faced by ﬁrms in the same industry. On the contrary,
Maher found that the use of interﬁrm RPE is rare in practice.
The goal of this study is to reconcile the discrepancy between the theory and practice
of interﬁrm RPE based on a new development of agency theory. Recently, the theory is
moving in the direction of understanding the effects of limited commitment by contracting
Along these lines, this paper investigates the value of monitoring information
in renegotiable contracts.
Allowing contracts to be renegotiable creates a cost of using
monitoring information that was ignored in Holmstr¨om’s (1979) analysis. When this cost
is taken into account, the sufﬁciency part of the informativeness criterion no longer holds.
Instead, whether a signal is useful in incentive contracting will depend on the quality of the
Corresponding address: Department of Accounting, School of Business and Management, Hong Kong University
of Science and Technology, Clear Water Bay, Hong Kong. Tel: (852) 2358-7558. Fax: (852) 2358-1693.