Review of Accounting Studies, 9, 97–116, 2004
# 2004 Kluwer Academic Publishers. Manufactured in The Netherlands.
Inter-Departmental Cost Allocation and Investment
DONNA WEI firstname.lastname@example.org
INSEAD, Boulevard de Constance, Fontainebleau 77305, France
Abstract. This paper endeavors to demonstrate that ﬁxed cost allocation can align investment incentives in
a multi-period and multi-division setting. In a decentralized ﬁrm, a divisional manager can make an
investment that beneﬁts both his own and the operations of a downstream division. The relative budgeted
activity (RBA) cost allocation method assigns ﬁxed cost charges according to the ratio of a division’s
budgeted activity in proportion to that of the ﬁrm, and thereby resolves the hold-up problem created by
the decentralized setting. Internal accounting rules can be designed to give managers strong incentives to
internalize the ﬁrm’s objective regarding efﬁcient investment levels, and alleviate the tension between ex
ante investment efﬁciency and ex post production efﬁciency. This paper examines how much the ﬁxed
charges should be in order to achieve the optimal level of investment.
Keywords: cost allocation, transfer pricing, incomplete contracts, depreciation
JEL Classiﬁcation: D92, M41
Despite its widespread practice, the rationale behind the allocation of ﬁxed costs has
not been fully understood.
Fixed costs are usually deﬁned as costs that do not vary
with changes in production volume, within a speciﬁed time period. By ﬁxed cost
allocations, I refer to the spreading of ﬁxed costs across periods and organizational
subunits. When ﬁxed costs only pertain to production within one division and to one
product for a single period, there is no real ‘‘allocation’’ issue. Critics of ﬁxed cost
allocation argue that such costs are irrelevant to subsequent decision-making once
the investment is made and become sunk.
As a consequence, the critics argue, the
spreading of these costs can lead to double marginalization which in turn causes
inefﬁcient production decisions.
On the other hand, Zimmerman (1997) points out that, ‘‘if common costs are not
allocated, the managers in the two divisions have less incentive to invest . . . the
optimal level of the common costs.’’ My model formalizes the intuition that the
upstream manager under-invests because he fails to internalize the beneﬁts of
the investment accruing to the downstream division. In the absence of cost sharing,
the upstream manager gets only a portion of the total beneﬁts, but bears the full cost
of the investment. The resulting discrepancy in goals between the division and the
ﬁrm gives rise to a hold-up problem frequently discussed in the economics literature.
In this analysis, I rationalize the allocation of ﬁxed costs both across divisions and
over time. The notion that ﬁxed lump-sums, when allocated independently of actual
output levels, can avoid any double marginalization problems is not new. However,
to the best of my knowledge, no prior research has examined how much these lump-
sum charges should be in order to align investment incentives. This paper introduces