Review of Accounting Studies, 3, 289–320 (1998)
1998 Kluwer Academic Publishers, Boston. Manufactured in The Netherlands.
Welfare Effects of Timely Reporting
MITCHELL A. FARLEE
Faculty of Commerce and Business Administration, University of British Columbia,
Vancouver, British Columbia, Canada V6T 1Z2
Abstract. A principal-agent model is examined in which a manager acquires private cost information sequentially.
All possible communication schemes are equivalent to one of two: (1) timely reporting, where the manager reports
as soon as possible, and (2) delayed reporting, where the manager delays the report of the ﬁrst of two signals. In
the primary case identiﬁed, timely reporting is shown to be “owner valuable.” However, the manager is better off
under delayed reporting. Finally, total expected surplus is shown greater under delayed reporting. The owner’s
beneﬁt from timely reporting is less than the manager’s loss.
Most sufﬁciently large organizations use a budgeting process. The primary purposes served
are planning, control, motivation, and coordination (Merchant, 1998). This paper studies a
budgeting setting where the ﬁrst three purposes are illustrated.
in advance) takes place through a budget-setting contractual process which determines
planned productionchoices. An ownerdesigns thecontract’s compensation plan to motivate
the manager to produce as planned. The contract is designed to exert control over the
manager’s consumption of information rents (which may be interpreted as budgetary slack).
Budgeting processesare quite naturally dynamic in nature (Anthony, Dearden, and Govin-
darajan, 1992; Merchant, 1998) involving negotiation and ﬂows of information throughout
the organization. Effective budgeting involves participation by subordinate managers. Par-
ticipation serves two primary purposes: (1) increased goal commitment, and (2) increased
information ﬂows (Anthony, Dearden, and Govindarajan, 1992). In this paper, the budget-
ing process is always designed to achieve goal commitment. The second issue, information
ﬂow, is the focus of attention.
In such a dynamic budgeting environment it seems natural to assume that earlier informa-
tion is unambiguously better, holding the cost of information production ﬁxed. This paper
challenges that statement. In the presence of private information and rent-sharing it is not
only possible but logically plausible that this assumption is false.
This paper uses a theoretical modelto study a budgetingsetting in whichan agent privately
acquires information over time. I assume two pieces of information are acquired by the
manager before producing. Because the manager acquires the information sequentially, the
manager may be asked by his superior (owner) to participate in the budget-setting process
by reporting his information in one of two ways: (1) timely reporting, in which the manager
reports information as soon as it is acquired, and (2) delayed reporting, where the manager
reports only after all information is acquired. Timely reporting provides the owner with
I use the model to study the welfare consequences of the alternative reporting schemes.