See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/331024975

Costing Systems

Article  in  Foundations and Trends® in Accounting · April 2019

CITATIONS READS
0 7,877

1 author:

Eva Labro
University of North Carolina at Chapel Hill
48 PUBLICATIONS   3,918 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Eva Labro on 26 April 2019.

The user has requested enhancement of the downloaded file.

 COSTING SYSTEMS

Eva Labro1
University of North-Carolina at Chapel Hill
Kenan-Flagler Business School
Eva_Labro@unc.edu

As published in Foundations & Trends in Accounting (2019), Vol 13, No 3-4, pp. 267-404.
http://dx.doi.org/10.1561/1400000058

Final accepted version February 2019

1
Thank you to the Foundations and Trends in Accounting series editors, Sunil Dutta and Stefan Reichelstein, to offer
me the opportunity to write this monograph. I wholeheartedly thank my frequent co-author on costing system topics,
Ramji Balakrishnan, for countless hours of discussion on all things “costing systems” over almost 15 years. Many of
the ideas and thoughts presented in this manuscript originated in our joint work and related discussions. I also thank
an anonymous referee, Vic Anand, Bart Dierynck and Stephen Hansen for comments.
 COSTING SYSTEMS

Abstract

This monograph provides a structured overview of costing system research that can explain the
variation in the characteristics and properties of costing systems found in practice based on firms’
source(s) of their demand for cost information. Costing systems are not developed in a vacuum
but are designed to fulfill a purpose. In order to have a meaningful decision on the various demands
for cost information, I start in Part A by exploring the different techniques firms can use to supply
cost information to its managers and employees. Next, I discuss how even the most advanced
costing systems will contain error, and present research on how the production environment affects
this costing error and what firms can do to reduce the error, outlining important avenues for future
research. Part B then moves onto the sources of the demands for cost information. I first discuss
the decision-making objective of costing systems, and the requirements it places on the preferred
measurement object, which resources to include in the cost measurement, and the desired
properties of the costing system. The most voluminous literature here is on the capacity acquisition
and allocation decision problem, but I cover the demands on costing systems to support customer
portfolio decisions, inventory management decisions and decisions on managing competition as
well. This part also explores the demands for costing systems for the identification of cost
management opportunities and inventory valuation to support financial and tax accounting in a
chapter each. Part B concludes with a chapter on cost information supporting performance
measurement and control, again with respect to the preferred measurement object, which resources
to include and the costing system’s preferred properties. Part C develops on the conflicting
demands placed on costing systems by firms’ attempts at designing costing systems that serve
multiple purposes. Furthermore, this part also discusses the nature of the strategic interactions
between those different demands.

Keywords

Costing systems, cost systems, managerial accounting system, cost system design, cost
information

1
Table of Contents
Table of Contents ....................................................................................................................... 2
1. Introduction ......................................................................................................................... 5
PART A: PRODUCTION OF COST INFORMATION ....................................................... 11
2. Techniques to produce cost information............................................................................. 11
2.1. Traditional costing systems......................................................................................... 12
2.2. Activity-Based costing systems .................................................................................. 15
2.3. Resource Consumption Accounting ............................................................................ 18
2.4. Time-driven costing systems ...................................................................................... 20
3. Errors in costing systems ................................................................................................... 24
3.1. Even advanced costing systems contain error.............................................................. 24
3.2. How can we improve costing techniques? ................................................................... 29
3.3. Production environment affects costing errors ............................................................ 30
4. Important remaining questions on errors in costing systems ............................................... 32
4.1. More realistic true production functions...................................................................... 33
4.2. Updating of costing systems ....................................................................................... 36
4.3. A life cycle approach to costing system properties ...................................................... 39
PART B: DEMANDS FOR COSTING SYSTEMS .............................................................. 40
5. Decision-making demands for costing systems: general takeaways .................................... 41
5.1. Measurement object and which resource costs to include for decision-making ............ 41
5.2. Short run and long run decisions ................................................................................. 42
5.3. Desirable properties of costing information for decision-making ................................ 44
6. Capacity acquisition, allocation and pricing decision demands for costing systems ............ 46
6.1. Measurement object and which resource costs to include for the capacity acquisition,
allocation and pricing decision .............................................................................................. 47
6.2. Desirable properties of costing information for the capacity acquisition, allocation and
pricing decision ..................................................................................................................... 53
6.2.1. Issues around error in costing systems ................................................................. 53
6.2.2. Ability to measure unused capacity...................................................................... 55
7. Other decision contexts’ demands for costing systems ....................................................... 56
7.1. Customer profitability analysis ................................................................................... 57
7.1.1. Measurement object and which resource costs to include for customer management
decisions ............................................................................................................................ 57

2
 7.1.2. Desirable properties of costing information for customer management decisions . 59
7.1.3. Interactions between customer and product/service costing.................................. 60
7.2. Inventory management decisions ................................................................................ 61
7.2.1. Measurement object and which resource costs to include for inventory management
decisions ............................................................................................................................ 61
7.2.2. Desirable properties of costing information for inventory management decisions 62
7.3. Competition management decisions: desirable properties of costing information ........ 63
8. Identification of cost management opportunities ................................................................ 65
8.1. Measurement object and which resource costs to include for identification of cost
management opportunities..................................................................................................... 65
8.2. Desirable properties of costing information for identification of cost management
opportunities ......................................................................................................................... 69
8.3. Interorganizational cost management .............................................................................. 70
9. Inventory valuation for financial and tax accounting .......................................................... 74
9.1. Measurement object and which resource costs to include for inventory valuation ........... 74
9.2. Desirable properties of costing information for inventory valuation ................................ 75
10. Performance measurement and control demand for costing systems ................................... 76
10.1. Measurement object and which resource costs to include for performance measurement
and control ............................................................................................................................ 76
10.2. Desirable properties of costing information for performance measurement and control. 79
10.2.1. Multiple performance measures ............................................................................. 80
10.2.2. Variance analysis and flexible budgeting................................................................ 82
10.3. Deliberate misallocation to circumvent performance measurement and control............. 82
PART C: INTERACTIONS AND CONFLICTS BETWEEN SOURCES OF DEMAND
FOR COST INFORMATION AND IMPLICATIONS FOR COSTING SYSTEM DESIGN
86
11. Interactions and conflicts between sources of demand for cost information ......................... 86
11.1. Conflicts ....................................................................................................................... 86
11.2. Interactions................................................................................................................... 88
12. Interactions and conflicts between demands on cost information for performance evaluation
and decision-making ................................................................................................................. 91
12.1. Conflicts ....................................................................................................................... 91
12.2. Interactions................................................................................................................... 93

3
13. Transfer Pricing: Interactions and conflicts between tax reporting and managerial accounting
demands on cost information .................................................................................................... 95
13.1. Conflicts ....................................................................................................................... 95
13.1.1. Management accounting focus ............................................................................... 95
13.1.2. Tax Accounting focus ............................................................................................. 97
13.1.3. One Set of Books or Two? ...................................................................................... 98
13.2. Interactions................................................................................................................... 99
14. Conflicts and interactions between demands for cost information for inventory valuation,
management and control ......................................................................................................... 101
14.1. Conflicts ..................................................................................................................... 101
14.2 Interactions.................................................................................................................. 102
15. Conclusion ...................................................................................................................... 103
16. References ....................................................................................................................... 105

4
 1. Introduction

Costing information is used in a myriad of organizational and managerial settings to make

decisions, manage costs, value inventory and/or evaluate performance. Virtually all companies use

some shape or form of costing systems to support these objectives. Of interest, large variety in the

characteristics and properties of costing systems exists, indicating that a “one size fits all”

perspective on costing systems is likely inaccurate. This monograph provides a structured

overview of costing system research to explain the co-existence of different costing practices in

the cross-section, accompanied with my commentary on this exciting research area, which I hope

will be of interest to academics. On multiple occasions, this body of research has come to

prescriptive conclusions, which I believe will also be valuable and insightful to practitioners

designing costing systems and managers using reported cost data.

While “cost” is a term commonly used in business and daily life, a very specific yet

generalizable definition does not exist, as such definition depends on the purpose or objective of

the cost measurement exercise. Hence, in this monograph, I will follow the very high level

definition by Demski and Feltham (1976) of cost as “a description of the sacrifice associated with

some abstract cost object”. A costing system, then, is defined as the set of procedures used to

measure cost. The premise is that firms have limited information about the underlying true cost

function they would like to approximate, and costing systems will help provide such

approximation. Costs are not measured in a vacuum, just for the sake of measuring them; the

objective of our purpose for this measurement needs to be considered. That is, we need to consider

the specific endogenous demand for cost reports (Demski 1981).2 I will explain how the choices

2
As the references in this introduction make clear, the idea of linking the type of information that the cost accounting
function is asked to provide with the objective for which the cost information will be used is a very old one. Clark
(1923, pp. 236) already lists 10 different objectives for the demand for cost accounting information and the required

5
of the “abstract object” that needs to be measured and which “sacrifices” to include in that

measurement depend on the source of the demand for cost information.

I structure this monograph around four primary purposes of such cost measurement

exercise. The first is decision-making, also termed decision-facilitating by Demski and Feltham

(1976). Firms make many decisions, such as capacity acquisition, capacity allocation, product mix,

pricing, supplier selection and closing of business units or plants. Under this objective, we consider

a single decision maker (e.g., firm or employee) optimizing a particular problem. In order to be

useful for decision-making, costing systems try to measure the marginal cost relevant to the

decision at hand. The second purpose is cost management. Firms typically want to do more with

fewer “sacrifices” in order to stay competitive. An understanding of the production processes will

be key here, as will be the amounts of resources available and used. Inventory valuation for

financial and tax accounting constitutes the third purpose. Here, compliance with regulatory

requirements drives demand for cost information. Cost management and inventory valuation can

be seen as special cases of a decision-making objective. For a cost management objective, cost

information will help facilitate the firm’s decisions about resources that affect the cost structure,

whereas for inventory valuation, cost information is provided to facilitate decisions by outside

stakeholders such as users of financial statements and tax authorities. Because the characteristics

of cost information useful for these two types of decisions are different from the types of decisions

studied under the heading of the first objective, I will treat them as different objectives. The last

purpose is control and performance measurement, also termed decision-influencing by Demski

and Feltham (1976). This introduces a multi-person context, with different people optimizing their

corresponding cost measurement. For readers without time to tackle this book’s 502 pages, I recommend Frank (1990)
who discusses the main insights from the book, demonstrating how many insights in contemporaneous management
accounting originated with Clark (1923). Vollmers (1996, Appendix B) lists the objectives for cost accounting
specified by a large set of textbooks published between 1921 and 1948.

6
own objectives from which they derive utility. Shareholders want to measure how well the CEO

and the firm’s leadership team are performing, managers aim to control the work of their

employees, and firms desire to understand the performance of outside parties, such as suppliers.

When viewed through an economics lens, we can couch these relations in a Principal-Agent

setting. In order to be useful for performance measurement, costing systems ought to provide

measures of whether the Agent spent time on the right tasks and worked hard or not.3

The purpose of the cost measurement will not only determine what object we want to

measure (e.g., marginal cost in decision-making settings) and which costs to include as “sacrifices”

(e.g., all factory costs for inventory valuation for financial accounting statements), but also what

constitute desirable properties of cost information, such as relevance, reliability, accuracy, level of

detail, timeliness, compatibility, up-to-datedness, and frequency of reporting. Following Clark

(1923, chapter IX)’s old adage “different costs for different purposes”, Kaplan (1988) demands

that costing system properties should differ for each objective. For example, Kaplan (1988) argues

that decision-making requires frequent detailed cost information, while for inventory valuation for

financial accounting and tax purposes, a firm needs much more aggregate and less timely cost

information. As another example, Clark (1923, chapter XII) argues that to “tie in” with financial

accounting, cost accounting needs to measure nothing more than total operating expenses, while

to detect losses, waste and pilfering, a firm needs complete records of actual costs and standards

for efficient production against which to compare these. Throughout this monograph, as I develop

3
Different classifications of objectives for managerial accounting exists. For example, as indicated, Demski and
Feltham (1976) put forward a coarser classification that distinguishes solely between decision-facilitating (decision-
making) and decision-influencing (performance measurement). Kaplan (1988) lumps the cost management objective
in with both the decision-making purpose (with cost reduction as a decision made by management) and the control
purpose (as a type of operational control). I prefer to separate cost management from both as (1) the “object” on which
to collect cost information is not marginal cost as in the other decision-making examples and (2) the control of costs
is exerted on an object, rather than on an agent who has incentives and their own objectives from which they derive
utility, which is key in how I have defined the performance measurement objective in its multi-person context.

7
on each cost measurement objective, I will define what constitutes the “abstract object” and

“sacrifices” for each purpose, and discuss the properties of cost measurement most desirable for

each objective.

Before we can have a meaningful discussion on the different demands placed on costing

systems, we have to first cover the techniques by which the cost information can be supplied. Part

A of the monograph relates to the production or supply choices made by costing system designers.

It covers the literature on the techniques typically used in the production of cost information in the

presence of limited information about true cost behavior, the literature on errors in costing systems

which researches how close the approximation by the costing system gets to that true costs, and

the unanswered questions in this literature. Part B moves onto the demand side for cost

information. First, I cover the first objective for which the provision of cost information is

requested: decision-making. The decision-making purpose of cost measurement is the most

developed in the academic literature, and over the decades has steadily grown in importance in

management accounting textbooks from being the subject of 6% of chapters in 1940s textbooks,

over 33% in 1960s textbooks (Horngren 1989), to over 60% in the latest edition of the leading

managerial accounting textbook by Horngren et al. (2015). After a chapter that covers the general

insights on costing to support decision-making, I move to more detailed insights for several

specific types of decisions. A lot of the work has focused on the capacity acquisition, allocation,

product mix and pricing decisions, but customer portfolio decisions, inventory management

decisions and decisions on managing competition are also important and discussed. Next, I look

at cost management, inventory valuation, performance measurement and the requirements these

demands place on the cost measurement object, the resource sacrifices to consider and the desirable

properties of the costing system.

8
 As will become abundantly clear from Part B, the different sources of demand for costing

information affect which costing system properties are most beneficial. Firms typically need cost

information to support many of these objectives, and hence face a problem. Practically, it is very

difficult to have more than one costing system, with every costing system designed to be optimal

for one specific objective, or even one specific decision. Not only is the expense of developing

and maintaining multiple costing systems prohibitive for many firms, but it would also result in

confusion among the users of the cost information, and may even get firms in trouble with the tax

authorities if they use different “books” for tax reporting purposes than for, say, divisional

performance evaluation. In part C, I develop on the conflicts created between different demands

for product costing. I contrast information for decision-making purposes with information more

useful for performance measurement in general. I will also cover issues that arise in transfer pricing

where conflicting requirements are placed on cost information with respect to measuring and

rewarding performance of business units, corporate tax minimization and the coordination of

production and procurement incentives. Next, I will contrast costing requirements for inventory

management, inventory control and inventory valuation. Lastly, I will cover exemplar research

papers that identify settings where different demands for cost information may, rather than conflict,

simply interact in a neutral way, and interactions may potentially be beneficial.

Before moving to part A, I want to outline the scope of the literature reviewed in this

monograph. My focus is on papers that specifically relate to cost information and the properties

thereof. Cost information is arguably one of the most important aspects of managerial accounting

information, and while the broader literature on properties of managerial accounting information

can also be relevant to the topic of cost information and I will refer to some select papers in this

literature, as a whole that literature is out of the scope of this monograph. Furthermore, my main

9
perspective is looking at costing systems through an economics lens, and this monograph surely

underrepresents research on the topic from a psychology or sociology (or any other) lens. I refer

the reader to the Foundations and Trends in Accounting piece by Luft and Shields (2009) for an

overview of psychology based research in management accounting.4 Next, this monograph aims

to be useful both to novices in the costing field who are searching for a primer on this literature as

well as for people familiar with (some of) the literature who are interested in a structured overview

and thoughts on where future research avenues may lead. Hence, given this heterogeneity, an

advanced reader can skip or skim some sections in the monograph.5 Lastly, while the reference list

of this monograph is very long, the overview is by no means complete nor exhaustive, and I

apologize upfront to the authors of relevant research papers that I have missed or left uncited.

4
While Luft and Shields (2009) is not specifically about costing systems, they feature in many sections in this work
(e.g. sections chapters 5 and 6). While I am not aware of an overview piece that specifically studies costing from a
behavioral perspective, other behavioral accounting research summaries that cover some of this work are Luft and
Shields (2003) and Sprinkle (2003).
5
For example, big parts of chapters 2, 5 and 10 are of a more basic level.

10
PART A: PRODUCTION OF COST INFORMATION

2. Techniques to produce cost information

Various techniques developed over time to produce cost information and supply managers

and employees with cost estimates. In this chapter, I will describe the core features of each method

using a common notation and discuss the most important differences between techniques. The

underlying premise of any costing system is that companies face limited information about true

cost behavior. Each of these costing methods, in essence, tries to estimate or approximate an

underlying true cost function that describes how cost changes with changes in volume or in the

level of an activity or process related to that cost (Labro 2006a). Cost objects are the products,

services, distribution channels, customers or any other part of the business for which a manager

may wish to understand how much of the firm’s resources it consumes. My focus is on the

techniques used for the allocation of indirect costs, as direct costs are assumed to be traced through

to the cost object of interest without error.6 This chapter borrows heavily from my prior work in

Balakrishnan et al. (2012a) and Hemmer and Labro (2017).

Different costing system techniques provide different answers to three core questions. First,

how do we chose which cost pools to form, and how many cost pools do we define? Cost pools

group the cost of distinct resources. Second, which cost driver or cost allocation basis do we use

to allocate the cost of these cost pools to cost objects? 7 Third, what is the denominator value used

6
Direct costs can be measured or traced through to the cost object without error, while indirect costs need to be
allocated as their consumption by the cost object is not measured. For example, when a manufacturing plant of a
company has an electricity meter at the point where electricity enters the building electricity costs are direct to the
cost object of the manufacturing plant, but indirect to the different product lines manufactured in this plant.
7
I use the terms cost drivers and allocation bases as synonyms. A cost driver is any factor, which causes a change in
the cost of an activity or cost object. In order to be able to be used as a cost driver, the consumption of the driver by
the activity or the cost object needs to be directly measured. Indirect costs can then be allocated to the activity or cost
object in the same proportion.

11
to construct the cost driver or cost allocation basis? I begin with some common notation. Consider

a firm with I capacity resources, K cost objects, and a Leontief production function.8 Each of I

resources supplies a certain amount of capacity at some cost per unit of capacity. Capacity is

measured for different resources in different units such as machine hours, labor time, and square

feet. Let RC1 … RCI be the costs of the resources with the total cost TC, given by TC = å RCi .
i

Each distinct resource forms its own pool, and firms typically use a large number of different

resources. Assume that the firm classifies the total cost into J cost pools. Let αij > 0 represent the

portion of resource i that is allocated to cost pool j with ∑% 𝛼$% ≤ 1. If this term equals 1, there is

no unused capacity of resource i, whereas a strict inequality indicates the presence of unused

capacity. Then, the cost in pool j is CPj = å a ij RCi for j = 1 to J. Many values of αij might be zero
i

as many resources might relate to one or just a few cost pools. This allocation represents the first

stage in a typical two-stage allocation system – the stage that allocated resource costs to cost pools.

Let βjk > 0 represent the portion of cost pool j that is allocated to cost object k. Then, with the

second stage of allocations (the stage that allocates cost pool costs to cost objects), cost allocated

to the cost object k is COk = å b jk CPj for k =1 to K, and with åb

k
jk £ 1.
j

2.1. Traditional costing systems

Traditional costing methods typically group costs in one or very few cost pools, and

estimate cost as a linear function of volume or a cost driver highly positively correlated with

volume such as direct machine hours or direct labor dollars (Horngren et al. 2014), using a two-

8
Here, I assume that the underlying true production function is linear and that input quantities are used in fixed
proportions that are not substitutable, because most observed costing systems are linear. In chapter 5, I will discuss
the difficulty of producing cost information when the underlying true production function is non-linear or exhibits
flexibility in inputs.

12
stage design. Costs that changed in a fixed proportion to changes in volume (or a cost driver highly

correlated with volume such as direct machine hours or direct labor hours) are considered variable

costs, while costs that stay constant when volume changes are considered fixed costs over this

volume range. Let CDjk be the number of cost driver units (e.g., units produced, machine hours or

labor dollars) that relate to the consumption of cost pool j by cost object k, typically a product.

Two choices for the denominator volume of the allocation basis are possible. In an actual costing

CD jk
system, we have b jk = . That is, we derive the allocation percentages as the ratio of the
å CD jk
k

driver units to cost object k to the total of the relevant driver unit at actual levels. A normal costing

CD jk
system replaces the denominator with the expected total or normal volume; b jk = . Similar
CD jkN

computations apply to the first stage as well.

Johnson and Kaplan (1987), Cooper and Kaplan (1987), Shank and Govindarajan (1988),

Miller and Vollmann (1985) and others have claimed that these traditional costing methods were

systematically distorting product costs, leading to wrong decisions on the basis of these costs. They

critiqued the simplicity of only considering costs to be either variable with volume or fixed and

disapproved of the exaggerated use of direct labor hours as an allocation base for the indirect costs

in a “new” (at the time) production environment where fewer hours of direct labor were used,

given the manufacturing industry’s shift to more automation, and the economy’s shift to the service

industry. Also, a bigger share of the costs in this “new” production environment was argued to be

indirect and therefore had to be allocated using some allocation base.9 They argued that picking

9
This seems self-evident in the service sector, where a large proportion of costs are in staff functions and support
roles. On the other hand, Kerremans et al. (1991) find that in the manufacturing sector, increasing automation increases
the proportion of direct costs.

13
the wrong allocation base in this setting has disastrous consequences. Early empirical research at

that time (Foster and Gupta 1990) using data from 37 facilities of an electronics company

continued to find that the strongest explanatory power rested with volume based drivers. However,

most empirical research papers after that provide evidence that indeed costs are not driven solely

by volume. Banker and Johnston (1993) finds evidence that operations-based cost drivers need to

complement volume drivers to explain variations in manufacturing overhead in the airline industry.

Dopuch and Gupta (1994) documents that changes in the production process such as production

and engineering order changes affect both direct and indirect manufacturing costs in a beverage

company. Anderson (1995) documents the impact of product mix heterogeneity on manufacturing

overhead costs. Banker et al. (1995) find that while volume does explain manufacturing overhead

in their sample of 32 manufacturing plants, most of the variation in overhead costs is explained by

measures of manufacturing transactions. Balakrishnan et al. (1996) document that the complexity

of care provided by Ontario hospitals has a significant influence on operating costs.

Traditional volume-based allocation systems have existed for many decades. Despite

considerable criticism about the usefulness of traditional volume-based allocation systems, they

are still surprisingly widespread. Surveys in the US manufacturing sector indicate that 35% of

firms use such traditional costing system (Hughes and Paulson Gjerde 2003) and that 28% use a

single plant-wide overhead rate (Krumwiede 1998). In the European manufacturing sector,

Brierley et al. (2001) list evidence that the percentage of manufacturing firms that use one blanket

overhead rate is high, yet varies across countries from 5% in Finland, over 26% in the UK (Drury

and Tayles 1994) to 51% in Ireland. This study also indicates that in these various European

countries, direct labor cost or direct labor hours is that cost allocation basis of choice in between

50 and 80% of cases. Also in the UK, but going beyond the manufacturing sector to include the

14
financial and commercial sectors, Al-Omiri and Drury (2007) find that 35% of their surveyed

sample uses a traditional costing system, with 5% using only one cost pool. Brierley (2008)

documents a median number of cost drivers in the UK of only one in 55 interviews, while Drury

and Tayles (2005) finds that 59% of the manufacturing firms in the same country use between one

and three cost drivers only. Lastly, Cinquini et al. (2013) documents the very extensive use of

direct labor as the main allocation basis in their survey of Italian manufacturing firms. However,

by necessity, these surveys tend to have small sample sizes, focus on specific regions, and are

mostly performed on manufacturing firms; thus, the extent to which these estimates are global

averages is unknown.

2.2. Activity-Based costing systems

As the criticism on traditional volume based costing systems grew, Activity-Based Costing

(ABC) was proposed as a more accurate costing method whereby both more cost pools are formed

and allocation bases are chosen to better reflect the cause and effect relationship in resource

consumption patterns at each cost pool (Cooper and Kaplan 1988, 1998b). ABC estimates changes

in cost as a function of changes in activity level, where an activity is any discrete task that an

organization undertakes to make or deliver a product or service (Bruns and Kaplan 1987).

Activities form cost pools, which may cut across departments, in contrast to the use of cost centers

(departments) in traditional costing. New cost drivers, other than volume-based drivers, are used

to allocate the cost of the resources aggregated in these activity cost pools. Examples include the

number of set-ups to allocate the cost of the set-up activity, the number of purchasing orders to

allocate the cost of the procurement activity, the number of machine insertions to allocate the cost

of the machining activity, the number of inspections to allocate the cost of the inspection activity,

and the number of different components to allocate the cost of maintenance of the Bill of Materials.

15
Furthermore, the set of cost objects to consider moves beyond just products, to also include

customers, suppliers, distribution channels and any other measurement object for which the firm

wants to obtain a cost estimate.

The final innovation of ABC is to introduce the ABC hierarchy10: an understanding that

costs are driven by (and hence variable with respect to) activities that occur at different levels,

rather than simply variable with volume (units) or fixed, as proposed by traditional costing

systems. The typical hierarchy considers 4 levels: unit, batch, product (or service)-sustaining and

facility-sustaining. The hierarchical level at which a particular cost is classified indicates when

this cost becomes variable. Costs on the unit level are the costs that are traditionally called variable

costs and are incurred per unit (e.g. price). Costs on the batch level are incurred each time a batch

is delivered or brought to the production line (e.g. inspection and set-up costs). Product-sustaining

costs are incurred to enable the production and sale of a particular product (e.g. product design and

product advertising). Facility-sustaining costs are costs that are fixed in the short run. They only

become variable when the facility is closed down or reduced in size. This ABC hierarchy helps

management identify which costs are incremental for different types of decisions. For example, if

the decision concerns whether or not to produce one extra unit of a product, only the unit level

costs (such as the material to use in the unit) are relevant. However, for the introduction of a new

service to the firm’s service mix, all costs up to the service-sustaining costs (such as service

development and service specific marketing) are to be considered.11

10
In German language, this innovation can be traced back to Agthe (1959).
11
Note that the concept of the ABC hierarchy provides more fineness to the notions of fixed and variable used in
traditional costing systems, where such fixity or variability is defined w.r.t. changes in volume. ABC terms such
variable costs unit level costs, as the idea is that they are incurred with each unit produced and/or sold. Fixed costs are
facility level costs. Traditional costing systems typically misclassify the intermediate levels in the hierarchy. Product-
sustaining costs are lumped in with the fixed costs, disregarding the fact that they are variable with higher-level
decisions in the firm, such as the expansion of the firm’s product offering. Batch-level costs are lumped in with
variable costs, ignoring that they are incurred no matter what the size of the batch (its volume) is.

16
 With ABC, a focus on measuring the cost of unused capacity also developed because the

technique advocates the use of practical capacity rather than budgeted capacity to derive the cost

rates used in the allocation (Cooper and Kaplan 1992). That is, practical capacity becomes the

denominator choice. Practical capacity is the amount of capacity that a firm can “practically”

consume, and is calculated as the maximum theoretically available capacity adjusted for scheduled

downtime, training, holidays, etc. In contrast, budgeted capacity (as used in the traditional costing

systems) is the amount of capacity that a firm plans to consume over a budgeting period, which is

usually one year. This choice of denominator is important, because the use of practical capacity

delinks the supply (defined by practical capacity) of and the demand (defined by planned

consumption) for capacity. With ABC, the amount of unused capacity at cost pool j can now be

P
,./ 𝑇𝑅%, where TR j is the practical capacity for cost pool j.
identified as 𝑇𝑅%* − ∑-

There is substantial survey evidence on the adoption of ABC, although penetration varies

considerably across countries. Early survey evidence in US manufacturing (Krumwiede 1998)

suggests 17% of the 225 US manufacturing respondents adopted ABC. The largest survey includes

2789 responses, and indicates that 26% of US manufacturing plants are extensive users of ABC in

1997 (Ittner et al. 2002). Two contemporaneous surveys point to somewhat different ABC

adoption rates. Cagwin and Bouwman (2002) identify 204 US internal auditors of which 23% use

ABC significantly, while 32% use it somewhat. Hughes and Paulson Gjerde (2003) find that 38%

of the 130 US manufacturing firms surveyed either use ABC or a combination of ABC with

traditional costing methods. Elsewhere, early evidence from 75 Norwegian firms shows a very

high adoption rate of 40% for ABC (Bjornenak 1997), while Italian evidence suggests that only

10% of 132 firms use ABC in 1996, which increases to (only) 18% by 2005 (Cinquini et al. 2013).

Surveying 584 CIMA members in Australia, New Zealand and the United Kingdom in 2007,

17
Askarany and Yazdifar (2012) identify ABC adoption rates between 36 and 42%, while in the

same year Al-Sayed and Dugdale (2016) identify 37% of 152 surveyed UK manufacturing firms

as users of ABC. Al-Omiri and Drury (2007), on the other hand, find that only 29% of 147 UK

firms are using ABC. Lastly, Schoute (2011) finds that Dutch firms surveyed in 2002 have only a

15% adoption rate for ABC, but that 25% of firms are in the process of adopting at that time.

There is also some inconclusive evidence that ABC penetration differs across industries.

Al-Omiri and Drury (2007) find that firms in the financial and commercial sectors have larger

adoption rates than firms in the manufacturing sector, in line with the early evidence by Innes et

al. (2000) that ABC adoption is higher in the UK financial sector. In contrast, Cagwin and

Bouwman (2002) and Cotton et al. (2003) find that ABC adoption in the manufacturing sector is

higher than in other sectors in the US and New Zealand, respectively. Of course, survey evidence

on costing system choices struggles with the fact that different respondents may have different

ideas on what constitutes an ABC system (Askarany and Yazdifar 2012), which may be causing

differential responses across studies. Additionally, and unfortunately, to the best of my knowledge,

more recent evidence of ABC adoption and use is lacking in the current decade. ABC systems tend

to be fairly costly to implement, given their extensive information needs. They are also difficult to

maintain. Every change to the activities performed in the organization theoretically necessitates

the re-calculation of all cost driver rates, since the allocations are based on percentages that need

to add up to 100%. This may potentially explain why adoption rates are not higher.

2.3. Resource Consumption Accounting

Resource Consumption Accounting (RCA) was developed by the cost management interest

group at the CAM-I consortium (Consortium for Advanced Management - International) and

explained in a series of articles by Anton van der Merwe in practitioner oriented journals (e.g. van

18
der Merwe and Keys 2002; Clinton and van der Merwe 2006). The RCA Institute has now codified

it at www.rcainstitute.org. RCA systems seek to blend the best features of ABC with features

borrowed from Grenzplankostenrechnung (GPK), a German cost accounting method with a high

level of granularity that was developed after World War II.12 RCA forms cost pools by grouping

resources together based on their “technology, skill and homogeneity.” Thus, the focus of the

system shifts from activities to resources. At each resource cost pool, RCA breaks out costs into

two buckets – fixed and proportional – and uses separate rates (and potentially separate drivers) to

allocate costs from each of these buckets. The fixed and proportional classification is preserved

for intermediate cost pools that combine costs from several resources. Each of these two kinds of

costs are allocated separately to the cost object (e.g., the product) from the intermediate cost pool.

Thus, we can distinguish between the fixed and proportional costs allocated to a cost object.

RCA employs theoretical, rather than practical, capacity to determine the denominator

volume to calculate the allocation rates for the fixed portion of resource costs (Clinton and Webber

2004). The idea is that managerial choices affect how much of this theoretical capacity is used for

production today. For instance, choosing to run only two shifts (letting the machine be idle for 8

of the 24 hours available in a day) is the outcome of a choice. RCA use of theoretical capacity

stresses the importance of the decision to control the level of committed resources (Tse and Gong

2009) and provides stretch targets in capacity management (Clinton and van der Merwe 2006). In

terms of cost driver choices, an output-based driver, causally related to the proportional cost, is

used to assign the proportional portion of resource costs. Rather than using allocations that are

based on percentages or dollar values, RCA prefers to use drivers that are operational quantities to

12
The core reference for GPK is the textbook by Kilger et al. (2002), which is written in German. Its first chapter was
translated to English and published in Management Accounting Quarterly (Kilger et al. 2004). Unfortunately, most of
the literature on GPK is written in German. Some English language references include Sharman and Vikas (2004) and
Krumwiede (2005).

19
reflect better the causal relation, as in ABC. Furthermore, RCA uses replacement costs rather than

historical costs to compute depreciations.

Despite these differences, RCA is still a two-stage allocation system. From a notational

perspective, we solely need to add superscripts for the resource costs and activity drivers to indicate

their variability with respect to the underlying driver. That is, we need to decompose RCi into
1231 5$6
𝑅𝐶$ and 𝑅𝐶$ , and preserve this distinction for the purpose of computing rates for activity

cost pools j. The proportion of cost pool j allocated to cost object k is then calculated as

TR jk 78,5$6
b Th
jk
, fix
= Th , fix
. Unused capacity at cost pool j is estimated as 𝑇𝑅% − ∑-
,./ 𝑇𝑅%, where
TR j

78,5$6
𝑇𝑅% is the theoretical capacity of the fixed resources allocated to pool j. There is no possibility

of unused capacity in the proportional cost. While there are some case studies published that use

this method (e.g. Clinton and Webber 2004; Al-Qady and El-Helbawy 2016), I am not aware of

any large scale survey evidence on RCA use.

2.4. Time-driven costing systems

Time-driven costing (TDC) was introduced by Robert Kaplan in 2004 in an article in the

Harvard Business Review (Kaplan and Anderson 2004), followed in 2007 by a book (Kaplan and

Anderson 2007). They argued that because ABC systems are so expensive to implement and

cumbersome to maintain, TDC, with its focus on time as the sole cost driver and its ease of

maintenance, should provide a viable alternative. Both ABC and RCA stress the notion that supply

and consumption of resources are distinct and that the cost of idle capacity should not be allocated

to production, but to the source of the demand for this idle capacity. For example, strategic reasons

such as planning for growth entail that the cost of idle capacity gets allocated as a period cost,

20
while the cost of unused capacity in the off-season gets allocated to customers in peak season.

TDC pushes these notions even further.

TDC begins by calculating a rate for each resource, with the calculation based on the

underlying practical capacity for the resource. The choice of the practical capacity metric (e.g.,

time or space) is critical as the denominator volume choice, and sets up a “price” for units of

various resources, such as the cost of an hour of warehouse employee time in a distribution center

or the cost of 10 square feet of operating theatre space in a hospital. In the next step, TDC

implementations develop time equations for each cost object in two steps: (1) express each activity

in terms of the quantity (measured in capacity units) of each resource required to execute one

transaction, and (2) record the number of transactions (for each activity) consumed by the cost

object. This is similar to choosing a transaction or duration driver in an ABC setting. These time

equations are based on workflow maps of the processes involved in the production of the product

or the delivery of the service. The time equation yields the total quantity of a given resource

consumed by a cost object. Determining product cost then only requires simple multiplication of

the rate and quantity for each resource, and summing over resources. Note that this implies that

the TDC system can easily adapt to changes to the activities performed in the company and to

changes in the denominator of the cost rate by adding an additional term in the time equation and

multiplying in with a different cost rate, respectively.

prac
Formally, let RCapi be the capacity associated with resource I, expressed in physical

units of time. Then, let hik be the total quantity of resource i consumed by cost object k, again

expressed in physical units. W aggregate the resources required for an activity in the time equations

directly by multiplying the number of transactions to determine the total quantity of resource i

consumed by cost object k to estimate hik. In this way, time equations allow us to determine the

21
proportion of resource costs allocated to the cost object directly without having to determine αij

(the first stage allocation proportion) and βjk (the second stage allocation proportion) individually.

prac
The rate per capacity unit is RCi / RCapi and the cost allocated is

(
COk = åhik RCi / RCapi
i
prac
). While this advantage of TDC has been emphasised in the
literature, Balakrishnan et al. (2018) are careful to point out that technically a first-stage design

decision still has to be made. That is, which costs should be pooled in the cost pool that is the

numerator in the calculation of the cost rate per minute? They argue that a the literature has

scarecely covered, in particular, the role of indirect costs incurred in supporting departments with

many TDC applications ignoring these costs or at most lumping them in the numerator of the cost

rate without providing further insight in what causes their use. Note too that, for each additionally

defined cost pool, separate time equations need to be written.13

These calculations facilitate the identification of idle capacity. Because at the end of the

reporting period, the total volume of cost objects k produced (denoted qk) is known, TDC systems

can at the point in time calculate the total quantity of each resource i used as ∑, 𝑞, h$, . Comparing

prac
this to the practical capacity supplied RCapi allows TDC to calculate the amount of idle

capacity. Multiplying that idle capacity amount of resource i with its rate per capacity unit (

prac
RCi / RCapi ) provides the total cost of unused capacity of resource i.

As far as I am aware, it is the most recent innovation in costing systems that has seen a

reasonable number of adoptions, with many case studies published (Kaplan and Anderson (2007),

Kaplan et al. (2013), Hoozée and Bruggeman (2010), Morratz and Lueg (2017), Scott et al. (2018),

13
Many of the early implementations of TDC lumped all resource costs in one cost pool only.

22
and the long list of case studies in Table 9.1 of Hoozée (2012). Many of these applications are in

the healthcare sector and initiated by Professor Bob Kaplan at Harvard Business School, with a

long reading list of cases published on https://www.isc.hbs.edu/health-care/vbhcd/Pages/TDABC-

Recommended-Readings.aspx14. Proponents of TDC wrote most of these cases, and, to the best

of my knowledge, survey evidence on its adoption is not yet available in the literature.15

Table 1 summarizes and compares traditional costing, ABC, TDC and RCA in terms of the

types of cost objects, the types of cost allocated and various characteristics of the allocation

procedure, such as the type and number of cost pools, the type and number of cost drivers and the

denominator volume used for calculating cost driver rates.

Table 1: Overview of characteristics of different costing methods

Traditional Activity-Based Time-driven Resource

costing costing costing Consumption
Accounting
Cost object Product/service Product/service, Product/service, Product/service,
customers, customers, customers,
suppliers, suppliers, suppliers,
distribution distribution distribution
channels, channels, channels,
batches, etc. batches, etc. batches, etc.
Type of cost Manufacturing All indirect costs. Primary focus is All indirect costs.
allocated overhead Includes on indirect costs Includes
manufacturing, of resources manufacturing,
SG&A, and pre- directly SG&A, and pre-
production consumed by production
overhead. cost object. Less overhead.
applicable for
allocating
overhead in
support
departments to
primary cost
objects such as

14
Website accessed on 8/23/2018.
15
See Hoozée and Hansen (2018) for a comparison between ABC and TDC.

23
 products and
customers.
Type of cost Usually Activities. Often One for each Grouped based
pools dovetailed with cuts across kind of resource on technology,
organizational departments. (with the same skill and
structure, e.g. Grouped cost rate) homogeneity
departments logically to form
business
processes.
Number of 1-few Many Corresponds to Most
cost pools the number of
resources.
Type of cost Volume or Anything for Primarily time. Anything for
driver something which usage by Theoretically, which usage by
highly correlated cost object or can employ other cost object or
with volume activity can be measures such as activity can be
(e.g., direct traced. space. traced.
labor, direct
material).
Number of 1-3 One driver for As determined by Two drivers for
kinds of cost each cost pool. time equation each cost pool
drivers Many different (one for fixed
drivers overall. costs, one for
Research proportional
indicates costs)
diminishing
benefits after 10-
15 kinds.
Denominator Budgeted Practical capacity Practical capacity Theoretical
for capacity capacity
computing
cost driver
rates

3. Errors in costing systems

3.1. Even advanced costing systems contain error

The previous Section lays out the big improvements that management accountants have

made to costing systems over time. Notwithstanding those, even the more advanced costing

systems are not error-free. In the early nineties, a literature ensued that researched errors in costing

24
systems, and the sources of these errors.16 This entire literature models linear costing systems, as

any cost system in practice applies such linear relationships to allocate costs. Noreen (1991)

derives three necessary and sufficient conditions for (ABC) product costs to reflect incremental

cost that capture the changes in cost as the result of a decision.17 First, total cost needs to be

partitioned into separable cost pools, each of which is only allowed to depend on one activity.

Second, within each cost pool, cost must be strictly proportional in a linear way to the level of

activity in that cost pool.18 Finally, activity drivers assigned to individual products can simply be

summed to arrive at total activity. Noreen (1991) explains how stringent these conditions are. First,

joint costs (where production costs are a non-separable function of the outputs of two or more

products, and unavoidable jointness arises from the technology used to produce the joint outputs)

can only be treated as incremental to all the cost objects which use the underlying resource, as they

have a public good characteristic. Examples are capacity for peak and off-peak demands, spare

capacity information technology, corporate advertising and corporate credit rankings. An oil

refinery takes crude oil and refines it into car gasoline, motor oil, kerosene, and heating oil in pre-

specified proportions. A decision to increase the production of kerosene will affect not just the

cost of kerosene, but will lead to increased production of all the other products that crude oil

transforms into as well. That is, the cost is incremental to the portfolio of these products, and not

just to kerosene.19 Furthermore, non-linear functions at cost pool level (e.g., quantity discounts)

are also ruled out as are linear functions with non-zero intercept at cost pool. This is especially

16
Section 3.1. draws heavily from Dierynck and Labro (2018).
17
See also Bromwich and Hong (1999) and Pereira and Mitchell (2012).
18
Noreen and Soderstrom (1994) present evidence from Washington hospitals that this condition does not hold.
19
Maher and Marais (1998) illustrate the problems identified by Noreen (1991) using simulations based on data
collected during a field study in a hospital. Trenchard and Dixon (2003) illustrate how the inability to accurately cost
joint cost in the not-for-profit manufacture of blood products inhibits managers in deciding on the optimal balance
between a joint platelet production model versus a non-joint alternative.

25
problematic since true cost functions are likely non-linear.20 Lastly, any interdependences between

products in the production process result in a violation of the necessary and sufficient conditions.

As one example, flexible manufacturing systems could create such interdependencies. While there

is some progress on the joint cost, non-linearities and interdependencies front (e.g., Hemmer 1996;

Christensen and Demski 1997; Anderson and Sedatole 2013) which we will discuss later in this

monograph, research on these important topics remains scarce.21

Datar and Gupta (1994) complements Noreen (1991) by pointing out different sources of

costing errors in the supply of cost information.22 They assume an underlying true cost function

that is linear, unlike Noreen (1991). The first error is specification error where a wrong cost driver

is chosen that does not exhibit a cause and effect relationship in resource use by products. For

example, the traditional costing systems are high on specification error because of an overuse of

volume-based allocations. The second error is aggregation error where similar cost drivers are

added together, and cost and units of a resource are aggregated over heterogeneous activities to

derive a single cost allocation rate. For example, marketing and insurance costs may be pooled

together in a “general overhead” cost pool. The demand for more refined costing systems arises

from the desire to reduce specification and aggregation errors. However, various types of

measurement error can be made when doing so. First, there is measurement error of cost pools

20
This begs the question why a firm would implement such linear costing technology if it knew its cost functions were
non-linear. We even lack anecdotal evidence of firms actually employing such systems. First, it may consider a linear
costing technology to offer a good enough approximation of the non-linear relation. Second, the complexity of a non-
linear costing system might be too high or make cost information hard to understand and use. Third, it may be
prohibitively expensive.
21
Some older work attempts to more normatively address the joint cost issue (e.g., Roth and Verrecchia 1979; Billera
et al. 1981; Balachandran and Ramakrishnan 1981, 1996; Biddle and Steinberg 1984; Williams and S. 1983; Cheng
and Manes 1992). Note that most of these works do not model an environment with informational limitations, where
the need for a costing system arises from a measurement or approximation objective as is the focus of this monograph,
but assume full information is available.
22
Note that this literature implicitly or explicitly assumes that direct costs are traced accurately, and that the errors are
incurred in the inaccurate allocation of the indirect costs.

26
where costs are wrongly identified costs with a particular cost pool. For example, financing costs

are accidentally added to the marketing cost pool. Note that financial accountants and auditors

focus on reducing this type of measurement error by ensuring each invoice is booked in the correct

general ledger account. Second, there exists measurement error of units of allocation bases, where

the specific units of resources consumed by individual products are wrongly measured. For

example, the reported usage of the set-up activity by a specific product line is higher than its true

usage. Note that all these errors are defined against an underlying true cost function that exhibits

linear relationships. In a series of examples, Datar and Gupta (1994) illustrate that one type of

error may over-cost a product, while another may under-cost a product, but that overall the product

may be fairly accurately costed because of the trade-offs between these errors. However, if the

firm refines its costing system by removing or reducing one error, while leaving the other error

intact, it may lose the trade-off effect that (partially) cancels out both errors. As a result, the firm

may end up with a less accurate product cost. Datar and Gupta (1994) conclude that partial

refinement of costing systems, an approach advocated by ABC advocates for reasons of reducing

resistance to change, limiting scope and cost of new systems implementations, etc., may not always

work.

Researching errors in costing systems is difficult as it requires that the researcher not only

observes the costing system in place, but also the underlying true cost against which to benchmark

the cost reports produced by the costing system. Therefore, Labro and Vanhoucke (2007) put

forward simulation (also termed “numerical experiments”) as a research method that allows

researchers to observe both, with the resulting errors defined as a function of the differences

between the benchmark true cost and the observed cost report. With the notable exception of

Christensen and Demski (1997) which I will discuss in Section 4.1., all research literature has

27
assumed that the underlying true cost benchmarks are generated by a linear functions. Therefore,

the existing academic literature is likely underestimating the error in the costing system, since in

practice true cost functions are likely to not be linear in many situations.

Following on the definitions of the various errors by Datar and Gupta (1994), in their

simulation study of a two-stage cost allocation system, Labro and Vanhoucke (2007) show that

the trade-off with its accompanying offsetting effect these authors warn for typically only happens

in very extreme cases, and that the only more widespread, potentially problematic, trade-off occurs

between measurement error on the resource cost drivers and aggregation error on the activity cost

pools. They also show that measurement error on the activity drivers impacts inaccuracy the most,

followed by aggregation error on the activity cost pools and that measurement error on the resource

cost pools impacts accuracy the least. Accuracy is measured using several metrics at the level of

the portfolio of products produced. The most commonly used metric is a Euclidian distance

measure of the difference between the true cost of each product in the portfolio and the costing

system’s approximation thereof. Other metrics include a mean percentage error across the portfolio

and the proportion of products that are costed “accurately enough”; that is, within a materiality

band of acceptable error. Cardinaels and Labro (2008) show in a laboratory experiment that also

in a TDC setting, people introduce measurement error in the time estimates used for the cost

calculations, even if they are incentivized to report as accurately as possible, and that this effect is

more severe when activities are more disaggregated, which is indicative of an endogenous trade-

off between aggregation and measurement error. Based on a laboratory experiment, Mastilak

(2011) finds that the classification of costs into cost pools affects the accuracy of individuals’

understanding of relations among costs, where they estimate relations between costs in the same

cost pool more accurately than those across different cost pools.

28
 3.2. How can we improve costing techniques?

Textbooks and practitioner articles have put forward several rules of thumb to improve the

accuracy of the cost information supplied by costing systems. For example, textbooks advocate to

focus accuracy improvement on situations where there is high diversity in resource consumption

patterns or to restrict attention the largest resources (also termed the “Willie Sutton” rule, after the

bank robber who, when finally captured, replied to the question on why he robbed banks as “that

is where the money is”). Unfortunately, many of these rules were being prescribed before research

into their efficacy was executed and often remain ill defined. Babad and Balachandran (1993)

develop an integer programming model to identify the number of cost drivers needed provide

guidance on how to select cost drivers to balance accuracy with information processing cost to

collect additional cost driver information. Homburg (2001) improves on this model. Hoozée et al.

(2012) study the trade-off between identification and estimation errors of estimated transaction

times in a TDC context, and investigate when adding terms to a time equation is justified. Hwang

et al. (1993) develop algorithms to choose the best allocation base at each cost pool.

In contrast to the prescription by the textbook rules of thumb, Labro and Vanhoucke (2008)

find, using a numerical experiment, that the diversity rule of thumb is not accurate when diversity

is defined as heterogeneity in resource sharing across the costing system, that the opposite is often

true, and that there exist important non-monotonicities in the relation between diversity and

accuracy. The rule is only accurate when diversity is defined as heterogeneity in the size of the

resource cost pools or in the proportional resource consumption at each cost pool. Balakrishnan et

al. (2011) demonstrate under which conditions size based-rules for forming cost pools outperform

correlation-based rules, and develop a blended method that combines both size- and correlation-

based rules of thumb that performs very well. Furthermore, they also study the effect of composite

29
cost drivers on accuracy, demonstrating the gains be made from including additional resources in

the cost driver index. Because numerical experiments are uncommon as a method in accounting

research, I refer to Anand et al. (2018, forthcoming) for an overview of the modeling blocks that

have been or can be used in research on costing systems, and for guidance on how to best execute

and report on research using this method.

Further research could also focus more on implementation issues in obtaining reasonably

accurate cost reports, as the cost of adoption and maintenance of costing systems is high.

Collecting the data necessary to develop a costing system and keeping it up to date constitutes a

major effort. Unfortunately, the academic accounting literature seems somewhat biased against

publishing work that gives strong prescriptive recommendations on how to improve management

accounting in general and costing in particular. Furthermore, action research, where researchers

get their hands dirty in companies (e.g. by implementing costing systems) to provide insights into

implementation and maintenance stumbling blocks is difficult to publish. As a result, most

guidance on implementation comes from practitioners. Employing a constructive case-study

research approach that balances the need for practical and academic contributions may provide a

solution (Kasanen et al. 1993; Labro and Tuomela 2003).

3.3. Production environment affects costing errors

The organization of the production environment can also affect the general accuracy of the

costing system. 23 Consider the distinction between a job shop organization and a process shop

organization. In a job shop, small batches of a variety of products are manufactured and most of

the products produced require a unique set-up and sequencing of steps. In a process shop, larger

23
Section 3.3. draws heavily on Hemmer and Labro (2017).

30
batches of more similar products are manufactured using a similar setup and common sequencing

of steps. These two ways of organizing production are typically considered the extreme ends of a

continuum. In a production organization that leans more toward a job shop there is little sharing

of resources across products, whereas a system that is closer to a process shop is characterized by

a lot of resource sharing with most products making use of the same set of resources, even if the

pattern of resource consumption varies across products. Balakrishnan et al. (2011) find that job

shops require a more sophisticated costing system with more cost pools than process shops to

achieve the same overall level of costing accuracy. The reason is the following: at every step in

the cost calculations, measurement error can be made. However some errors may offset resulting

in a fairly accurate approximation of the cost of the cost object (Datar and Gupta 1994). The

likelihood that costing errors offset each other is much higher in a process shop because various

products share resources, which means that over- and under-allocations are more likely to cancel

out. In a job shop, however, the likelihood that errors offset each other is much smaller, and hence

a more sophisticated costing system is needed to achieve the same level of accuracy.

The production environment also endogenously affects the accuracy with which costs can

be measured, potentially constraining the accuracy of the cost information that can be supplied.

For example, for the ABC hierarchy to be valid, the setup cost of a single batch should be

independent of the number of units produced in that batch, and hence not depend on the size (in

terms of volume of units) of the batch. However, Economic Order Quantity (EOQ) rules used in

the production environment may introduce associations among these two cost hierarchy categories

(Ittner et al. 1997). In their simplest form, EOQ inventory policies calculate optimal order quantity

?@A
in units as 𝑄< = > B
, whereby D is demands in units over the time period, S is the ordering cost

per order in dollars and H is the inventory holding cost per unit over the time period. That is, the

31
optimal batch size 𝑄< is a function of trade-offs between order costs (which are incurred at batch

level) and inventory holding costs (which are incurred at unit level). Hence, EOQ rules used by

the procurement function entail that the number of batches is positively associated with production

volume since the total number of batches represents the product’s production volume divided by

the constant reorder point. This invalidates the assumption of the ABC hierarchy that the batch

and unit level costs are independent. Such endogenous production scheduling based on EOQ

models thus makes variation in costs associated with production batches empirically

indistinguishable from variation due to production volume (Anderson and Sedatole 2013).

4. Important remaining questions on errors in costing systems

Several research questions remain unanswered in the literature on errors in costing systems.

Below, I develop on two areas where I see many opportunities for research contributions. First,

stepping away from simple linear underlying true production functions for the costing system to

approximate is difficult, but would be an enormous step forward to increase the practical relevance

of the research in this area. As mentioned in Section 3.1., existing research likely underestimates

the errors in cost systems because it assumes that linear costing systems approximate linear true

production functions. Real-life production functions are most likely frequently non-linear.

Furthermore, firms can opt for flexible technologies that allow for substitution between inputs and

potentially create interdependencies between resources. Second, most research so far has taken a

static snapshot view of costing systems, whereas the firm’s environment is constantly evolving

and changing over time, which warrants studying issues created by a more dynamic view of costing

systems, such as the updating of these systems. Of course, the topics discussed here are non-

32
exhaustive, and I refer to the last section in Anand et al. (2018, forthcoming) for a list of additional

fruitful areas of research on costing errors.24

4.1. More realistic true production functions

Most of the papers on errors in costing systems assume that the underlying production

technology that costing systems are trying to approximate is Leontief with constant returns to scale,

meaning that each product uses resources in fixed proportions and substitutability among resources

is impossible. This is convenient from a modeling perspective, because it means reported costs can

be presented as a linear additive function (Christensen and Demski 1997), which corresponds to

the linear features of costing system designs.25 Hence, this simplifying modeling choice can be

motivated by the prevalence in practice of linear costing systems. However, realistic production

functions are likely non-linear and exhibit scale economies (Christensen 2010). Furthermore, the

opportunity to buy flexible manufacturing technology (often at a premium) indicates that firms

value technologies that allow for strategic substitution of resources (e.g. as input prices change or

capacity constraints arise) and/or even of outputs (as demand changes). Christensen and Demski

(1997) show that even if the underlying production function remains Leontief, introducing scale

24
Service department allocations are another interesting avenue. Management accounting textbooks typically cover
three methods to deal with interdepartmental service allocations: the direct method that ignores allocations to other
service departments, the sequential step method, and the recommended reciprocal method that sets up a set of
simultaneous equations. However, the most recent available evidence (Drury and Tayles 1994) (which admittedly is
dated) indicates that only 21% of companies even set up separate department overhead rates, while 27% employ a
single plant-wide rate for all overhead (including support costs). 45% allocate support overhead costs to production
departments, and charge the production department costs by means of a departmental overhead rate. Hence, there
seems to be some discrepancy between the textbook recommendations and practice. Unfortunately, academic research
on the topic seems to have mostly stopped by the nineties (e.g., Jacobs and Ronald 1987; Balachandran et al. 1987;
Atkinson 1987; Lambert and Larcker 1989; MacArthur 1998; Jacobs and Marshall 1999).
25
Kaplan and Welam (1974) is an early paper that presents a linear programming overhead allocation formulation for
a non-linear production function, albeit in the appendix to the paper.

33
(dis)economies may create errors in product costs, and that on occasion simpler (e.g. direct labor

based) allocations may be preferred.26

Next, considering a Cobb-Douglas production function where the input factors are

substitutes and the rate of substitution between input factors changes (unlike in a linear production

function where it remains constant), Christensen and Demski (1997)’s simulation analysis suggests

that the interaction effect of costing procedures and technology becomes even more complicated.

Error and bias are likely to arise, even though it is difficult to analytically specify where in the

portfolio of products they may be most problematic. Once moving to flexible technologies where

inputs can be substitutes, we are in difficult territory, and I believe it would be useful if the

numerical experiment analysis could be extended to incorporate this feature to provide more

generalizable insights.27 Hence, the two extremes of no substitutability (Leontief) or infinite

substitutability (Cobb-Douglas) have been discussed. I am not aware of any accounting literature

that has covered the intermediate case of partial substitutability.

To address scale economies in numerical experiments, some thought needs to be put into

what constitutes the best proxy for error in the costing system. Most numerical experiment studies

keep the set of cost objects the same for comparability reasons as total resources can then be held

constant, but that ignores the fact that firms with scale economies will grow demand to use them,

and hence does not seem suitable as a modeling choice if the research question relates to the effect

of scale economies. Furthermore, it may be the case that with changing volumes, bias in the error

is more important than solely inaccuracy. For example, overcosting may be more problematic than

26
See also Christensen and Demski (2003).
27
Dhavale (2007) models the error made by using a fixed proportion cost allocation method when the underlying
production technology has variable proportions and inputs are potentially substitutable. I will discuss this paper in
Chapter 6, as he does so in the context of a pricing decision which is covered in that chapter.

34
undercosting as the growth strategy may be wrongly questioned.28 TDC has some potential to

incorporate non-linearities because the time equations on which the technique is based can be

viewed as conditioning statements which approximate a non-linear cost function with piece-wise

linear approximations (Balakrishnan et al. 2012b).

Several field studies have documented costing problems that arise in practice in settings

where technologies are flexible, such as when flexible manufacturing systems and computer

integrated manufacturing is used (Dhavale 1989; Goose 1993; Datar et al. 1993; Anderson and

Sedatole 2013). These studies can provide input into which features of these settings are the prime

candidates to be modeled and manipulated in numerical experiments, so that practical advice can

be generated. For example, an interesting question arises from the tension in Anderson and

Sedatole (2013). In flexible manufacturing settings, costs of product variety such as setup time

(typically a batch level cost) and engineering support (typically a product level cost) shift to facility

level, because procurement of the specific type of machinery minimizes these types of costs. Does

this restore the relevance of traditional costing systems that simply distinguish between fixed and

variable cost, and consequently make ABC systems less relevant again? Firms that are making no

investments in flexible technologies and continue to rely on manual set-ups (the costs of which are

measured at batch-level) are likely to see a strong association between product variety and batch-

level costs (Ittner et al. 1997). In contrast, firms that have flexible automation systems are unlikely

to observe such association between breadth of product offering and batch-related costs. However,

28
A starting point may be found in Anand et al. (2017) that uses a decision context with a profitability measure, rather
than a costing error metric, and that also models the resource consumption matrix in units of inputs consumed by units
of outputs rather than percentages. When volume changes, it will be important to think about resource consumption
in units rather than in percentages.

35
in such firms, batch level and facility-level costs are negatively correlated, again invalidating the

assumption of independence of levels in the ABC hierarchy (Hemmer and Labro 2017).

Furthermore, if firms endogenously respond to the need for offering product variety by

adopting flexible automation systems, their cost measurement system may underestimate the

resulting costs since a regression analysis will not document any batch cost level variation (as these

are close to zero) nor any fixed investment cost variation (by definition, since these costs are fixed)

that is associated with an increase in product variety (Hemmer and Labro 2017). Anderson and

Sedatole (2013)’s study concludes that ABC does not necessarily become irrelevant under this

type of flexible production technology, but that the granularity and frequency with which data are

obtained to estimate these cost relationships may contribute to the researchers’ inability to establish

costs behave in line with the ABC cost hierarchy. A numerical experiment could tease out which

features of the costing system design are most likely affected by a change to a flexible technology.

4.2. Updating of costing systems

A costing system provides a static view of the firm’s resource consumption by cost objects,

even though firms operate in environments that are subject to exogenous change, while their choice

of production technology, which the costing system is trying to reflect, and product mix may also

endogenously change over time. Such changes are bound to increase the error in the snapshot

provided by the costing system, and hence practitioners advocate for firms to regularly update their

costing systems (Horst 2013a, 2013b; West 2013). Indeed, one of the selling points used by TDC

advocates is that it is much easier to update a TDC system than, say, an ABC system (Kaplan and

Anderson 2004). When unions negotiate a higher wage rate, the cost rate that is multiplied into the

time equation can simply be increased. When clients demand an additional activity such as a

quality check, the minutes it takes to do such a check multiplied with the number of checks can

36
easily be added to the time equation. On the other hand, each of these type of updates would require

all percentage allocations to all activities to be revisited in an ABC system.

Given the practical importance, the literature on either the frequency or the thoroughness

with which a costing system update should be performed is surprisingly limited. The practitioner

literature, again, remains vague: “revise as frequently as necessary to keep information accurate”,

“update when the market has changed dramatically”, “update your system when it is antiquated”,

“go without a revision if your operations and cost structure are stable”, and “overhaul your system

when its assumptions are no longer true” (Horst 2013a, 2013b; West 2013). “Update cost rates

annually” seems to be the most explicit advice given (Horst 2013a). The academic literature is

virtually silent, so there are lots of unanswered questions. What exactly does a costing system

update entail, as options range from an update of overhead rates to doing a full overhaul of the

entire costing system, whereby also the grouping of resources into cost pools is revisited? What is

the benefit to a costing system update relative to doing nothing and sticking to potentially stale

cost information? Is there a difference in the robustness of the costing systems to different types

of changes (e.g. input price changes versus input quantity changes)? Does the direction of change

matter (e.g. is the firm improving efficiency or reducing it)? Are simpler systems with fewer cost

pools more stable than more complex multi-pool systems? Do any of the responses to the above

questions change when the nature of the change is endogenous, in that a cost management and

efficiency improvement exercise can be driven by the prior snapshot cost report produced by the

costing system? As far as I am aware, Anand et al. (2014) is the only paper that begins to address

a small subset of these questions. Using a numerical experiment, they find that the benefits of

partial updates (e.g., updating overhead rates) are limited and the gains to a thorough update (e.g.,

regrouping resources into cost pools) increase in the magnitude of the change. The implication is

37
that a periodic thorough overhaul dominates frequent incremental changes. Minor updates can be

an overreaction to small changes in efficiency. Many more questions are left unanswered, and I

believe this is a particularly fruitful opportunity for doing practically relevant research.

Empirical research that describes costing system updating patterns over time is also limited,

but could bring awareness about what firms are or are not doing to maintain the quality of their

cost information. Most research is cross-sectional rather than longitudinal and has focused on

costing system first-time adoption rather than updating (e.g., Davila and Foster 2005; Chenhall

and Morris 1986; Maiga and Jacobs 2008; Ittner et al. 2002). Since the majority of firms have

adopted costing systems by now, understanding what determines updates and what the benefits of

updating are becomes increasingly important. Using the HIMMS (Healthcare Information and

Management Systems Society) Analytics Database® of hospital information technology that

covers yearly surveys of the near-census of hospitals in the US between 1987 and 2010, Labro and

Stice-Lawrence (2019) find hospitals update their AS in response to three types of pressures. First,

economic pressures such as demand side increases in the need for information driven by state-

level price regulations and supply side increases in the quality of accounting information driven

by vendor rollouts of improved AS; second, coercive pressures imposed by regulators mandating

certain practices, such as internal control practices imposed by Sarbanes-Oxley Section 404; and

third, mimetic pressures for hospitals to conform their AS to those of their peers, such as local

county and prominent “celebrity” peers. They find that only economically driven updates lead to

economic benefits in the form of lower operating expenses and higher revenues, suggesting that

costing updates help identify low hanging fruit for cost management opportunities in this sector.

In contrast, they find some evidence that AS updates prompted by coercive regulatory pressures

actually impose economic costs in the form of higher operating expenses. It would be interesting

38
to examine effects of costing system updates on additional outcome variables to gain more insight

into how exactly such operating expense reductions are obtained, and address the generalizability

of these results in other industries.

4.3. A life cycle approach to costing system properties

Prior research documents that organizational needs for management accounting systems

change over time, as the firm proceeds in its life cycle. Moores and Yuen (2001) document that

firms demand more formality in their management accounting systems as they age. In the

healthcare sector, Labro and Stice-Lawrence (2019) find that hospitals first adopt the basic

financial transaction systems of accounts payable, general ledger and patient billing, followed by

case mix systems. Subsequently, typically costing and budgeting and credit/collections systems

are adopted. Finally, executive information systems penetrate hospitals. Davila and Foster (2007)

find that startups in the information technology and biotech sectors typically first adopt financial

planning systems, followed by human resource and strategic planning systems, while financial

evaluation systems are adopted at a later stage. While none of these papers study the detail of the

costing system features longitudinally29, particularly the latter study suggest that different

objectives are important at different stages in a firm’s life cycle, with decision-making support

being focused on early, while performance evaluation objectives follow later. It would be

interesting to study if costing system properties evolve accordingly, shifting over the firm’s life

cycle from more useful to decision-making to more characteristic of a performance measurement

objective.

29
As discussed earlier, the majority of empirical research on this topic is cross-sectionally.

39
PART B: DEMANDS FOR COSTING SYSTEMS

In the preceding part A, I have hinted at the context in which costing systems are developed

and reported costs are supplied. Indeed, as explained in the introduction, costing is not done in a

vacuum but has a purpose. In this part B, I start with the demand for costing systems to support

decision-making because most research linking costing system design to an endogenous reason for

the existence of the costing system has been decision-making related. Chapter 5 will develop on

the role of cost information and its desired properties in a decision-making setting more generally.

Of course, people in organizations make many possible decisions. What the desired object of

measurement is, and what the resources to be included in the allocation and properties of the cost

information are, may depend on exactly which type of decision the cost information needs to

support.30

Chapter 6 makes the decision context much more specific, by discussing the work done on

the relation between costing systems and the capacity acquisition, planning and pricing decision,

which constitutes the brunt of the work done in the area of how cost information supports decision-

making. Chapter 7 ponders on other decision contexts such as customer portfolio management,

inventory management, and competitive decisions. For each of these decisions, I will discuss what

the cost object is that the costing system attempts to measure (the “measurement object”), which

resource costs are included or excluded in this measurement, and what the desirable properties of

cost information are for the specific objective.

While the literature on the decision-making demand for cost information and the costing

systems that support the production of such information seems most developed, as outlined in the

30
In fact, the plentitude of decisions firms are making was the “excuse” of the early numerical experiment literature
on costing system design to not single out one decision as more relevant, and hence work in a context-free setting.

40
introduction, there are at least three other sources of demand for cost information which I discuss

next. Firms also develop costing systems to identify cost management opportunities (Chapter 8),

value inventory to support financial and tax accounting (Chapter 9), and support performance

measurement and control (Chapter 10).

5. Decision-making demands for costing systems: general takeaways

5.1. Measurement object and which resource costs to include for decision-making

According to economic theory, we should make decisions by equating marginal revenues

with marginal costs, both of which hence need to be estimated. For any choice we have to choose

C DEFE5$G C H3LG
such that C H83$HE IJ2$JDKE
= C H83$HE IJ2$JDKE , as those are the relevant costs since they are the

opportunity costs of the resources used (Labro 2006a).31 Variable costs most closely approximate

such marginal costs. Furthermore, this prescription from economics means that sunk costs are not

relevant, and many fixed costs fall into this category.32 Practice, however, seems to behave

inconsistent with this economic theory. First, given so many different decisions are made in firms

on a daily basis, costing systems would need to report the marginal costs for each of these different

decisions. However, since costing systems are costly to implement, developing and maintaining

multiple costing systems is impractical. Second, we observe the extensive use of full costing,

whereby fixed costs are allocated (Drury and Tayles 1994; Brierley et al. 2001). As a consequence

of this wide gap between theory and practice of costing, costing practice seemed to evolve for a

long time without academic guidance. However, accounting research developed to reconcile the

31
Note that the scope of this monograph is limited to cost information, and does not cover other non-financial
information.
32
Of course, the opportunity costs of these fixed resources are still relevant to decision-making.

41
theory’s prescription of the use of marginal costs with the practice of full costing, narrowing the

gap again.33

First, ABC advocates argued that in the long-term all costs are variable and that (at least

for long-term decisions) all costs (i.e. full costs) should be included in decision-making (Cooper

and Kaplan 1992).34 Second, and specific to the capacity acquisition and planning decision, others

have shown that full-cost based pricing and capacity planning can be used as heuristics where the

cost allocations are valuable because they approximate the opportunity cost of the resource based

on its expected use at the time of resource acquisition (Labro 2006a). I will discuss this literature

at length in Chapter 6. Third, agency theory and its focus on incentives and performance

measurement can also explain the use of full costing. For example, Zimmerman (1979) shows how

cost allocations can create optimal incentives, while Kanodia and Dickhaut (1989) show that

private information and reputation building entail that sunk costs are taken into account. Hemmer

(1996) goes as far as deriving specific cost allocation methods that are optimal for providing

incentives. I will discuss this literature further in Chapter 10. Lastly, the behavioral literature has

researched factors that lead to the inclusion of sunk costs in decision-making, such as loss aversion,

reputational concerns, justification needs, group polarization, diminished personal responsibility

in group decision-making, etc. For an example of this literature, see Buchheit (2004).

5.2. Short run and long run decisions

The time horizon over which decisions are made determines which costs can be considered

variable over that period and hence should be included in reported costs to support decision-

making as relevant costs. This issue was already recognized by Clark (1923, Chapter IX) in his set

33
Some early literature also attempted this reconciliation (e.g., Lere 1986; Hilton et al. 1988).
34
Note that this is at odds with the definition from the Economics discipline that defines the long run as that time span
in which all costs become variable.

42
of examples on decisions taken at a car manufacturing plant. In his examples, the cost of producing

a car calculated to support a decision on temporary closure of the plant was substantially lower

than the same cost number calculated to help make the decision on shutting down the plant

permanently. Because the latter is a long-term decision while the first is only short-term, many

more costs are relevant to the permanent decision and can be altered in the long run.

Of course, this difference between decisions made in the short versus the long run has long

since been formalized and extensively covered in management accounting textbooks and courses.

We teach students that the contribution margin, which is price minus variable cost, is the relevant

concept to consider in the short run. As long as the contribution margin of a business opportunity

is positive and unused capacity is available, in the short run, the firm will do well to work that

business opportunity as it will make a positive contribution to the fixed costs of this capacity

already in place. If capacity is fully used, in the short run, firms will compare the contribution

margins of the different business opportunities and allocate the capacity to the one with the highest

contribution margin. A traditional costing system that disentangles fixed from variable costs seems

sufficient in this case.

In the long run many more types of costs become relevant as they can be altered. In this

case, costing techniques like ABC become very useful since they provide more insights on the

variability of costs that a traditional costing system typically considers fixed. For example,

marketing and product development costs are typically considered fixed in traditional systems, yet

the ABC method recognizes that these are variable at the product level in the hierarchy, which

needs to be incorporated in more long run decisions on contracting or expanding the firm’s product

portfolio. When the firm decides against launching an additional product line, it does not need to

43
assemble a design team to work out the products features and specifications, nor does the firm

need to create a product line specific marketing campaign.

5.3. Desirable properties of costing information for decision-making

Returning to economic theory, Blackwell’s theorem puts fineness forward as the criterion

for what makes good information for decision-making under risk when information is costless.

The idea is that finer (more accurate, in practitioner terms) information is better, if it improves the

predictability of the future state of the world x, as the decision maker can then adjust his decision

to match the course of action better to the likely state x that will obtain. Hence, it is crucial to

understand if the finer partitioning of the state space will make the decision maker update their

beliefs about the likelihood that x will materialize. The idea is that observing a finer piece of cost

information y will allow the decision maker to recompute his subjective probability distribution

over the state space, which is driven by the difference between the probability p(x) and the

conditional probability given observation of y, p(x|y). In such decision-making setting and in the

case where finer information is costless, the value of finer cost information can hence be positive

(if beliefs are updated) or neutral (if beliefs remain the same), but never negative as the decision

maker can ignore the information if he so wishes. 35

Of course, in practice, more accurate costing systems are not free. Producing cost

information, and designing and updating costing systems is costly, and firms can only choose

costing systems that are within their financial constraints. For example, Al-Omiri and Drury (2007)

provide evidence that smaller firms are less likely to have ABC or TDC systems because of the

35
Note that the call is for accurate actual costs, not accurate standard costs. Kaplan (1988) argues that standard costs
which are estimated or predetermined costs of performing an operation or producing a good or service under normal
conditions, are not useful to make decisions. Also, standard costs typically only include factory and manufacturing
costs, and not resources consumed to design, market and deliver the product which are also relevant to many decisions.

44
investment that is needed to set up such costing systems. Furthermore, even without financial

constraints, it is very unlikely that trading off the costs and benefits of developing more accurate

cost information results in firms opting for the highest level of accuracy achievable. From some

point onwards, the additional monetary costs to develop more accurate cost information outweigh

the benefits that can be monetarily proven (Dierynck and Labro 2018).36

If we also include behavioral considerations, an additional factor may limit the level of

accuracy in costing information provided: because people are boundedly rational, they have

difficulty processing vast and detailed amounts of information (Dierynck and Labro 2018). A

costing system that uses hundreds of cost drivers (rather than, say, 10) to allocate costs reports

information is harder to process by the decision-maker, and as a result decision-making may be

impaired. A stream of behavioral research studies these processing capabilities. For example,

Cardinaels (2008) documents that the impact of more detailed accounting information on better

decision-making quality depends on the match between the accounting knowledge of the decision-

maker and the format in which the detailed accounting information is presented. Furthermore,

given such bounded rationality, firms may choose to only introduce cost drivers that are of focal

interest (Merchant and Shields 1993). Lastly, King and Gupta (1997) document experimentally

that subjects can overcome the problem of having inaccurate costing systems by learning from

profit feedback, suggesting there exists alternate ways to obtain an understanding of costs not

directly supplied by the costing system.

36
As far as I know, there is no large-scale survey evidence available on the price of advanced costing systems in firms.
Anecdotally, prices seem to vary from a few $1,000 to millions depending on the characteristics of the firm and the
sophistication of the system (Dierynck and Labro 2018). Balakrishnan et al. (2011)’s evidence suggests that the
accuracy benefits of improved costing system sophistication taper off fairly quickly, so for firms with the least
sophisticated systems it is possible that the benefits do not outweigh the cost.

45
 As a result, when the cost of the increased accuracy in the cost report is factored in,

Blackwell theorem no longer provides a clean ranking on the value of this information in decision-

making, particularly since the benefit of improved decision making based on more accurate

information might be outweighed by the cost of more accurate information. Furthermore, in many

decision-making settings, more accuracy may not even lead to more decision-making benefits. For

example, consider a firm that must decide on accepting a special order at a fixed price 𝑝 but it does

not know its production cost. Suppose that the true production cost is 𝑐 the cost system reports a

cost of 𝑐̂ ≠ 𝑐. Clearly, accurate cost information can only have value for the firm if 𝑐
Q < 𝑝 < 𝑐

or 𝑐
Q > 𝑝 > 𝑐. In the former case, the firm could avoid to wrongly accept an unprofitable order

and in the latter case it could avoid to wrongly reject a profitable order. However, if 𝑝 > {𝑐, 𝑐̂ } or

if 𝑝 < {𝑐, 𝑐̂ } the firm cannot benefit from more precise cost information because it can identify

profitable and unprofitable orders despite the error in the cost system. Therefore, the literature

studies the specific decision-making context (the demand-side) to provide more insights into the

value of improved cost information supply. The next sections hence move on to discussing several

specific decision settings.

6. Capacity acquisition, allocation and pricing decision demands for costing systems

The capacity acquisition, allocation and pricing decision is the most commonly studied

decision in terms of its demand implications for cost allocations. More recent work has also

incorporated costing system errors in the context of this decision, which opens up new issues.

46
 6.1. Measurement object and which resource costs to include for the capacity

acquisition, allocation and pricing decision

The capacity acquisition, allocation and pricing decision has both a long run and short run

aspect. In the long run, firms need to decide how much capacity to acquire, which products to

produce and where to set preliminary list prices, which link the demand for resources (the solution

to the product planning problem) and the supply of resources (the solution to the capacity planning

and acquisition problem). In the short run, once more precise demand information has become

available, firms decide on the tactical prices, which can be different from the list prices, and on

how to allocate the available capacity among products. Balakrishnan and Sivaramakrishnan (2002)

explicate how all these decisions are an exercise in managing opportunity costs. Hence, we would

like that the costing system estimates the opportunity costs of resources consumed by cost objects,

which the literature mostly considers to be the products (or potential products) that form the

product mix. At the time of capacity acquisition, opportunity cost is the cash spent to acquire the

resource. Once acquired, opportunity cost is a function of the resource’s alternative uses, which

may vary over time. Because product demand is uncertain at the time of the capacity purchase,

firms may experience idle capacity (with opportunity cost equal to zero) in some periods and

shortages in others (with opportunity cost equal to the lost contribution). The optimal solution to

the capacity problem, then, trades off the opportunity cost of acquisition with the expected

opportunity cost of installed capacity given demand uncertainty.

The sharing of resources across products, where scarce resources allocated to one product

cannot be used to produce another product, severely complicates this decision problem. If only

one resource has inadequate capacity, we can solve the problem by rationing the resource to the

product with the lowest contribution margin. When multiple resources face binding capacity

47
constraints, the problem is more complicated, as the firm needs to identify the product mix that

minimizes total opportunity costs. Balakrishnan and Sivaramakrishnan (2002) develop a

framework in which to consider the role of cost allocations for these decisions, presenting a series

of models whereby they relax one assumption at a time. These models give a clear insight into the

research question offered in Section 5.1.: when are full-cost based allocations (where also fixed

sunk costs are allocated, contrary to the prescription from economic theory) appropriate for this

important decision, and when are they not? The series of models culminates in what Balakrishnan

and Sivaramakrishnan (2002) call the “grand program”:

𝑀𝑎𝑥*Z[,\],^][ 𝐸[a[a(𝑃,G − 𝑣, )(𝐴, +∈,G − 𝐵, 𝑃,G ) − a 𝜃$ 𝑐$ 𝑅$G ]] − 𝑇 a 𝑐$ 𝐿$

G , $ $

s.t.

a 𝑚,$ (𝐴, +∈,G − 𝐵, 𝑃,G ) − 𝐿$ − 𝑅$G ≤ 0 ∀$,G

,

𝐴, +∈,G − 𝐵, 𝑃,G ≥ 0 ∀,,G

𝑃,G ≥ 0 ∀,,G

whereby

§ N products (indexed k), M resources (indexed i)

§ Vk: variable costs per unit of product k

§ mki: units of capacity resource i consumed by one unit of product k37

37
Note that the accounting literature has almost exclusively worked with production functions without substitutability
between inputs, as discussed in Section 4.1. In the context of the capacity planning decisions, however, the Operations
Management literature has ventured further afield, and has also studied the flexibility of technology (e.g. Anupindi
and Jiang 2008; Goyal and Netessine 2007, 2011). This literature has the potential to help Accounting researchers to
make progress on this front.

48
 § Dk: demand for product k

§ Pk: price of product k

§ Li: units of resource i installed in the first stage, lasting T periods38

§ Demand function D = Ak – BkPk

§ ∈,G : demand shock

§ Ri: emergency capacity acquisition for resource i

§ ci: unit cost of resource i when bought at the time of capacity planning

§ Ѳi > 1: premium price to pay for emergency capacity acquisition

Of importance, note that this “grand program” is solved at product portfolio level, even in

the case where demand for products is independent like in this model. The connection across

products arises from the capacity constraints, because products share capacity resources. The

desirability of emergency capacity purchases depends on whether or not the resource constraints

are hard or soft. With soft capacity constraints, emergency capacity can still be procured in the

second stage, but typically at a premium Ѳi. When Ѳi is low, it is economically viable to augment

capacity to meet demand.39 With hard capacity constraints, the first stage is the only point at which

capacity can be procured. Technically, the premium is Ѳi=∞ for the hard constraints case, and so

it is never worthwhile to augment capacity. Hence, it could be the case that insufficient capacity

is procured to fulfill all product demand. Even though already subject to substantial simplification,

38
Another difference between the Accounting literature and the Operations literature is that Accounting literature
typically assumes that Qi is both the amount produced and sold of product i. Hence, there is no inventory. The presence
of an inventory of products would complicate the costing problem substantially, and assumptions would need to be
made about its longevity. Again, this stands in stark contrast to the vast literature in Operations on inventory
management (e.g. Feng et al. 2015a).
39
Even with soft capacity constraints it is possible that resources can only be acquired in a lumpy fashion rather than
continuously. The literature typically does not consider this case.

49
this problem is informationally and computationally demanding as the firm needs to anticipate

each possible demand state for its entire product portfolio at the time of capacity planning

(Balakrishnan and Sivaramakrishnan 2002). The literature identifies important features that

determine whether or not full product costs are an accurate approximation of the opportunity cost

that needs to be used in this capacity acquisition and allocation and list price setting problem, and

hence whether this portfolio problem can be decomposed in product level problems as the full

costing approach does.40

Whether or not tactical prices in the second stage can be set that are different from the list

prices set in the first stage affects the role for full costing. If prices are set in the first stage and

cannot be revisited in the second stage, tactical prices equal list prices and there is no updating of

(tactical) prices based on the arrival of new demand information. In this case, there is a list pricing

problem and a capacity planning problem, but no tactical pricing problem. The opportunity cost

(at time t=0) of overinvestment in capacity is the acquisition costs, and the opportunity cost of

underinvestment is determined based on the actual demand realizations, given we have to ration

capacity among the products. Hence, the opportunity cost of underinvestment at time t=0 is a

function of all possible linear programs rationing capacity in future periods. Because the list price

cannot be changed in the second stage, the full costing calculation will be relevant to calculate the

opportunity cost for the capacity planning problem under soft capacity constraints. However, when

capacity constraints are hard, this problem does not decompose into resource level problems,

because the opportunity cost of understocking capacity equals the lost contribution of unfilled

40
Part of the below summary overview is drawn from Dierynck and Labro (2018). For some exceptions and a more
nuanced discussion, we refer to Banker et al. (2002), Balakrishnan and Sivaramakrishnan (2002) and Banker and
Hansen (2002). Note that the bulk of the accounting literature on this topic uses analytical modeling methods. Hsu
(2011) uses hospital data to study the effect of cost allocations on pricing and is hence a notable exception.

50
demand, which is endogenous. Hence, full costing does not work to approximate that opportunity

cost.

If list prices can be altered after additional demand information arrives, there exists an

additional tactical pricing problem. With respect to the tactical pricing problem, the original

opportunity cost of capacity is sunk and not relevant, and the opportunity cost of capacity for this

tactical pricing problem solely depends on the available capacity and the demand shocks and can

be captured in a linear program. For the capacity planning problem at time 0, the opportunity cost

of underinvestment is the expected opportunity cost which is a function of all possible linear

allocation programs in the future. If capacity constraints are hard, this program too does not

decompose into resource level problems. Furthermore, the list pricing problem needs to be solved

at the product portfolio level rather than at the individual product level, as tactical pricing

determines both the lost sales and the contribution margin for sold products. This entails that full

costing, which considers each product independently, will not usually work to approximate the

opportunity costs when firms have such pricing flexibility.

Banker and Hughes (1994) is the core reference in this literature that established an

economic environment where using full cost in decision-making can be justified theoretically: the

case where the firm is list price setter facing soft capacity constraints. Interestingly, they develop

this result in a decentralized setting where extensive information is costly to communicate. Full

product cost can become a sufficient statistic for the communication of a more extensive

information set. If the marketing department makes the pricing decision while the production

department acquires capacity, these two decisions can be coordinated at no loss compared to a

centralized decision made by headquarters when based on full product cost. That is, the accounting

department only has to communicate full product cost, rather than all the underlying information

51
that goes into its calculation. This may explain why in practice, and as covered in Section 5.1.,

researchers have observed ample evidence of the use of full costing.41

As explained in Dierynck and Labro (2018), subsequently, the literature goes on to

establish the robustness of this result. Göx (2002) finds that the distinction between hard and soft

capacity constraints becomes moot in a one-resource model, and that the relevance of using

traceable (full) cost is solely decided by the firm’s ability (or lack thereof) to set tactical prices

after new demand information becomes available. The literature also proceeds to quantify the

magnitude of the economic loss when decomposing of full costing is used in situations where it

should not be (e.g. Balachandran et al. 1997). If the loss is not too high, firms may benefit from

employing a heuristic that is less informationally and cognitively demanding, rather than

attempting to solve the problem at portfolio level. As expected, the loss associated with full cost

based capacity acquisition and pricing will be smaller if a firm’s flexibility in emergency capacity

acquisition increases and its ability to set tactical prices decreases. Balachandran et al. (1997) study

the efficiency of cost-based decision rules for capacity planning. They model a setting in which

product cost data are used to infer the expected cost of under-and over-stocking and to determine

installed capacity. Comparing to a benchmark where there is perfect information on product

demand over the entire planning horizon, they consider the performance of various product- and

resource-based planning rules, and, using simulations, determine the conditions under which each

of these cost-based planning rules comes closest to the benchmark solution.42

41
Full-cost allocations have also been rationalized in a pricing context without a capacity acquisition decision (Ray
and Gramlich 2016). Additionally, incentive frictions can also explain the role for full costing in the capacity
acquisition decision. See, for example, the labor market frictions in Balakrishnan and DeJong (1993) and the literature
on capacity acquisition and allocation in multidivisional firms discussed in Section 12.2.
42
Other papers in the literature stream include (not exhaustively) Balakrishnan and Sivaramakrishnan (2001), Banker
and Hansen (2002), Banker et al. (2002), Dhavale (2005), Hansen and Magee (1993) and Balakrishnan and
Sivaramakrishnan (1996).

52
 6.2. Desirable properties of costing information for the capacity acquisition, allocation

and pricing decision

6.2.1. Issues around error in costing systems

All of the above literature on the role of full costing in the capacity acquisition, allocation

and pricing decision has been developed in a context where cost allocations are error-free

(Balakrishnan and Sivaramakrishnan 2002). Dhavale (2007) incorporates the error created by the

linear fixed proportion costing approximation where the underlying production technology

consumes inputs in variable proportions. In such variable proportion technology, different

combinations of inputs can achieve the same amount of output. The relevance to the capacity

allocation and pricing problem is that this technology allows the firm to substitute the unused

resource for the resource with scarce capacity in order to increase profits, even if such substitution

can become costlier as substitution progresses because production functions have a diminishing

rate of marginal substitution.43 In a model where the firm is a list price setter in a noncompetitive

market and faces soft capacity constraints, Dhavale (2007) shows that the conventional full costs

calculated under the fixed proportion technology need to be adjusted by a factor that is a function

of returns to scale of the variable proportion technology. His simulation results, then, compare a

Cobb Douglas production function with an approximation by a Leontief technology and suggest

that the resulting output prices set contain substantial error, which varies non-linearly in the returns

to scale of the technology.

43
Dhavale (2007) discusses two reasons why fixed proportion costing technologies may be applied in practice, even
if the underlying production technology exhibits the potential for input substitution. First, as discussed in Section 4.2.,
most costing research takes a static view. We can view the fixed proportion costing technology as a static case of the
variable proportion technology if input markets are competitive and price and demand parameters remain constant
during this short decision horizon. Second, the decision to change the way in which production is organized (e.g.
substituting manual labor with machine inputs) is usually a high-level decision (e.g. taken by top management), which
means that at lower levels in the organization, it is relevant to consider these input proportions fixed.

53
 Incorporating aggregation, specification and measurement error in this decision context

also substantially complicates the problem.44 Anand et al. (2017) marry the literature on product

planning decisions with the literature on costing errors surveyed in Chapter 3. Even in a world

devoid of uncertainty and under the stark assumption that capacity is acquired as needed without

a premium price, making these product planning decisions under limited information results in a

dynamic problem with a feedback loop (Anand et al. 2017).45 The outputs of a costing system, cost

estimates, contain error because of the limitations in the information on which they rely. When a

product mix decision is made based on these estimated reported costs, real capacity acquisition

occurs to produce the product mix. Resource expenditures are triggered and recorded by the

financial accounting system. The financial accounting system data on resource expenditures is an

input to the costing system, which allocates these costs to the products. Hence, these actual

expenditures will trigger a revision in the product costs. Rational decision-makers at the firm in a

deterministic world without uncertainty will, upon viewing these updated costs, review the past

decision. The decision should be self-confirming: the firm should not wish to change its decision

after updating costs, i.e., the system should be informationally consistent. Hence, the decision-

makers will choose to revise their product mix decision if it is inconsistent with the current reported

product costs. That is, even in the absence of uncertainty, a variance between reported cost in

period t=0 and period t=1 will obtain, and a dynamic process will ensue. Anand et al. (2017) show

that this problem can be modeled as a discrete nonlinear and non-analytic dynamical system and

define an informationally consistent equilibrium as one where the decision and the reported

product costs are self-confirming. They show that the first-best solution with full (unlimited and

44
This paragraph is drawn from Dierynck and Labro (2018).
45
Hwang et al. (1993) is an early paper that considers a decision context for a costing problem, but ignores this
feedback loop.

54
accurate) information is hardly ever an equilibrium solution. They devise a heuristic solution that

converges to an informationally consistent equilibrium solution quickly and efficiently (when it

exists) in the limited information case. This heuristic will allow researchers to conduct further

work that relaxes some assumptions on the capacity acquisition and allocation problem under

imperfect information on cost while being able to generate informationally consistent equilibria.46

6.2.2. Ability to measure unused capacity

As explained in Section 6.1., in order to accurately identify the opportunity cost of the

resources used, it is of great importance to know whether the resource has idle capacity, and if so,

how much. Indeed, idle capacity has an opportunity cost of zero in the capacity allocation and

pricing decision, while constrained capacity has a positive opportunity cost. One of the big selling

points of TDC advocates is that, as explained in Section 2.4, this costing method is better suited to

identify idle capacity, while other costing methods tend to under-estimate idle capacity or are even

unable to identify idle capacity. The argument underlying such statements is mostly one of a

behavioral response driven by bounded rationality. TDC uses a minutes or hours response mode

to understand how much of the resource (expressed in units of time) is typically used to perform

one unit of an activity, which, when multiplied with the volume of activity and compared to the

total units of time available for productive work (practical capacity) identifies the amount of idle

capacity. Other costing methods such as ABC typically ask the time use question in a percentage

response mode. The ability for TDC to better identify idle capacity stems from the behavioral

response of people to ensure that all percentages stated add to 100% to hide idle time. Furthermore,

46
For example, Homburg et al. (2017) additionally allow the firm to set its prices, rather than take prices from a
competitive market as in Anand et al. (2017), but do not incorporate the feedback loop identified in Anand et al.
(2017). It would be interesting to see simulation results for more complex decision problems that are based on
informationally consistent equilibria.

55
TDC respondents must be assumed to be boundedly rational and not multiply the volume of

activity with the number of minutes stated, as when doing so, they could theoretically adjust the

minutes responses to ensure no idle time is detected. In a laboratory experiment, Cardinaels and

Labro (2008) however, find that participants tend to overestimate the total time they spend on

activities by on average 37%, even in a setting where participants are paid to deliver the most

accurate estimates and no incentives to hide idle time exist. This result questions the ability of

TDC to accurately measure idle capacity. 47 More research on developing costing methods that

accurately identify idle capacity seems appropriate.48

7. Other decision contexts’ demands for costing systems

While the capacity acquisition, allocation and pricing decision is by far the most thoroughly

researched decision in management accounting as it relates to the role of costing information, it is

really just one of many sets of decisions firms take. This Section will cover three other decision

contexts in which we can think about specific needs for costing information: customer profitability

analysis, inventory management and decisions taken with an eye on managing the competition.

Even then, there are many more decisions that require some form of cost information that I do not

cover in this monograph such as project continuation decisions (Brüggen and Luft 2016),

production choices under cost-based reimbursement (Christensen and Demski 2003), and

production scheduling (Leitch et al. 2005; O'Brien and Sivaramakrishnan 1996).

47
Note that TDC advocates point not only to the ability of the costing method to identify idle capacity, but also to its
use for costing that idle capacity. Note that this aspect, theoretically, is irrelevant to the capacity allocation and pricing
problem since any idle capacity has opportunity cost zero here.
48
Potentially promising here is the research by Hasija et al. (2010) who develop a method to estimate the capacity of
contact center employees using aggregated historical data that have been distorted both by constraints on work
availability and by incentives for the employees to slow down when true capacity exceeds demand. That is, their
method undoes the bias in the aggregate information provided.

56
 7.1. Customer profitability analysis

7.1.1. Measurement object and which resource costs to include for customer

management decisions

Customer profitability analysis generally takes the customer as the cost object, and aims to

understand the different demands on activities customers generate and analyze their cost-to-serve.

While any measurement object selected to understand resource consumption will be customer-

centric, many options exist: customer costs can be measured at the transaction unit, over a time

period that ranges from one year to a lifetime (Foster and Gupta 1994), or focus solely on the

acquisition costs at the start of the customer’s relationship with the firm (Bjornerak and Helgesen

2012). Customer costs may even be measured before the customer begins a relationship with the

firm, as understanding customer needs and their willingness to pay is of utmost importance in the

product or service design stage, where management accountants want to understand what the cost

is of providing different levels of product or service attributes (Bjornerak and Helgesen 2012).

Costing techniques like ABC and TDC are particularly well suited to understand the

heterogeneity in cost-to-serve across customers, particularly if the ABC hierarchy is adapted to

focus on the demand side of the value chain by including customer-level costs and distribution

channel level costs. For example, including a distribution channel level in the ABC hierarchy will

provide insight to support a firm’s decision to run both a direct customer sales channel and a

channel that works via a distributor, as two different marketing campaign costs will be incurred.

Such cost analyses allow firms to calculate the profitability of each customer or each customer

segment which groups customers according to particular characteristics and gain insight into what

drives the high or low costs to serve (groups of) customers. Insights from these analyses typically

are that customers who order in small lot sizes or last minute, require customer specific features

57
and high service levels, pay late, and are unpredictable in their ordering pattern entail a high cost-

to serve (Balakrishnan et al. 2008). Such analyses also allow firms to rank customers from high to

low profitability, or even loss making, resulting in a whale curve of customer profitability: a

limited set of customers are highly profitable, the next set are more or less break-even customers,

while the last group of customers may lose money for the firm. From here, these cost reports can

support customer management and mix decisions.49

Two aspects complicate such customer profitability analyses. First, as suggested by the

study of Campbell and Frei (2010) in the context of the introduction of online banking services,

the outcomes of adopting a lower cost to serve channel are not necessarily always reducing the

total cost to serve, and could hence be counterintuitive. Campbell and Frei (2010) find that not

only the marginal cost of running an interaction through the online channel reduces for the bank,

but also the marginal cost for the customer of interacting with the bank decreases. As a result,

customers start to interact more with the online channel, generating additional service demands,

and (temporarily) increasing total cost-to serve. Hence, this study suggests that understanding the

cost impact on the customer side is equally important as understanding the cost effect on the firm

side. Second, marketing costs are often discretionary. That is, firm management decides how much

to spend on marketing through different channels (Foster and Gupta 1994). Management

accounting is notoriously bad at dealing with such discretionary costs, because as they are not

engineered, it is much harder to identify causal relations (Foster and Gupta 1994).50

49
Cugini et al. (2007) calculate the cost of customer satisfaction in the hospitality industry. High cost-to-serve
customers should not necessarily be dropped, because of knock on effects this may have on other aspects of the firm’s
business. Rather, managing these customers toward profitability should be the first concern. Subramanian et al. (2014)
additionally present some results where, counterintuitively, the high cost-to-serve customers are of strategic
importance in a competitive setting.
50
In contrast, manufacturing costs are much easier for costing systems to allocate because the underlying production
technology, which the costing system attempts to approximate, is clearly engineered. For example, even though I

58
 While the management accounting literature has made considerable progress on the cross-

sectional analysis of customer profitability, it has made much less progress on measuring the

heterogeneity of cost-to-serve and customer profitability over time, even as Foster and Gupta

(1994) called for such research. Casas-Arce et al. (2017) illustrate that such customer lifetime

value (CLV) metrics are important to improve decision-making in loan decisions by bank branch

managers.51 Interestingly, their results suggest that shorter tenure branch managers benefit more

strongly from the availability of the CLV metric, as it can substitute for their lack of experience.

Furthermore, Shin et al. (2012) show analytically that analyzing whale curves dynamically (over

time) introduces additional complexities. For example, price discrimination against a high cost-to-

serve customer may make such customer defer purchases. Notwithstanding their importance, it

seems that the management accounting literature has left the design of CLV measurement models

to marketing researchers. The many measurement models proposed in the marketing literature are

summarized in table 1 of Kumar (2018). One issue with this marketing literature, though, is that

all these models are prediction models based on data that firms happen to have collected, so they

are necessarily specific to each firm’s different context and data availability (Borle et al. 2008).

7.1.2. Desirable properties of costing information for customer management decisions

In a survey of 50 marketing executives, Foster and Gupta (1994) find that there is a big gap

in the perceived current ability and the potential ability of the accounting system to support their

decision-making. Marketing executives perceive the support for pricing decisions to create the

biggest gap, with the inability to allocate the total marketing budget to the individual marketing

know very little about car mechanics, I know a car is produced with 4 tires and a steering wheel, whereas I may be
less clear on the relation between marketing expense inputs and the sales volume generated by advertising these cars.
51
Mintz and Currim (2013) describe the results of a survey of 439 marketing managers, to identify determinants of
the use of financial metrics (many survey items refer to customer cost metrics, including CLV) and how their use
affects marketing-mix performance, overall finding a positive influence.

59
categories following as a close second. The third biggest gap relates to decisions on the percentage

of the total costs that ought to be devoted to marketing outlays. As the prior Section outlined, in

order to support customer management decisions it is important to slice the cost data in different

ways (different cost objects and levels of analysis, cross-sectional versus longitudinal). Not

surprisingly then, Foster and Gupta (1994)’s survey suggests that lack of flexibility in the costing

system is a chief complaint. Lack of timeliness and reliability, and not having enough detailed

information to disentangle fixed from variable costs also feature on the complaint list as properties

lacking in available cost information.52 The frequency by which customer profitability information

can be updated, however, can be somewhat lower than that by which product profitability

information is updated, given that slightly longer timespans are necessary to measure customer

cost and profitability and to capture enough of the cyclicality and recurring purchases of each

customer. This is in line with empirical evidence from the survey by Drury and Tayles (2006) in

the UK which finds that product and service profitability calculations are made more frequently

than customer profitability calculations.

7.1.3. Interactions between customer and product/service costing

Firms typically want to understand both customer and product or service profitability to

support their decision-making, as a one dimensional view of either customer cost or product cost

does not offer enough insights. Take the example of the use of a banking service like a checking

account by different types of customers. Some customers execute most of their transactions online

and hardly ever get cash out of the ATM, so the marginal cost of serving such a customer is small.

Others go into the branch for each transaction, using expensive time of the branch personnel, yet

52
Note that an aggregate measure of customer value at firm level is also of interest to equity market’s decision-making
(Bonacchi et al. 2015).

60
others rely heavily on cash and access to a vast ATM network. Average checking account usage

costs will not demonstrate these differences in usage by customers. As a solution, firms report

costs in a matrix view where product and customer costs are interacted.

Elaborating on the banking example, even if cost information is disaggregated by customer

type, it may look as if the checking account service is loss-making if costs are higher than revenues

across all customer types. However, banks would most likely be making a mistake by dropping

the checking account service altogether. Checking accounts are often the entry product for a

customer to establish a relationship with a bank. Hence, a more longitudinal view is necessary in

the cost reports. Indeed, it is likely unprofitable to serve a high school or even college student with

a checking account, but these clients take out much more profitable mortgages and credit cards

later in their lifetime, and may also open investment accounts with the same bank. The costing

system needs to represent that customer lifetime across the entire product portfolio view in order

for the bank managers to make good customer and product mix decisions.

7.2. Inventory management decisions

7.2.1. Measurement object and which resource costs to include for inventory

management decisions

For the support of inventory management decisions, it is important to understand accurately

which procurement related costs are at order (batch) level versus at unit (volume) level, as ordering

policies trade off both cost levels. Typically, ordering in larger volumes entails price discounts (at

unit level), less frequent order cost (at order/batch level), and less frequent reception and inspection

cost (at order/batch level), while inventory holding costs (at unit level) increase. Complicating the

accurate identification of these different resource costs is the fact that the use of economic order

quantity policies endogenously creates correlations between the different levels in the ABC

61
hierarchy (Ittner et al. 1997), an issue previously discussed in Section 3.3. Furthermore, it is most

useful if the aggregation level of the data coincides with the periodicity of the variation in

production activities (Anderson and Sedatole 2013).

In the long run, it is important to also include component level costs in the ABC hierarchy

when deciding on inventory policies. Component commonality, the use of the same version of a

component across multiple products (Fisher et al. 1999), increases the opportunities to save on

ordering and inventory holding costs as fewer (different) components need to be procured, while

allowing for procurement in larger batch sizes. Using common components also reduces risk in a

stochastic demand environment as, by pooling risk, the total volume of the common component

necessary in the production of multiple end-products can be forecasted more accurately (Baker et

al. 1986). The unit price paid for a common component, however, could be higher than that of a

unique component because a common component has more functionalities. Typically, component-

level costs such as component-specific research and development are higher than for unique

components as more design effort goes into a component that needs to be able to fit in multiple

products. For a complete and detailed analysis of the cost rate and cost driver effects of the use of

common components, I refer the reader to Labro (2004).

7.2.2. Desirable properties of costing information for inventory management decisions

Timeliness of cost information is particularly important for inventory management

decisions. Delayed information might cause delayed ordering decisions, which ultimately could

lead to a costly disruption in the production processes when the required materials or components

are unavailable when needed. 53 Bad inventory accounting processes and controls are costly to

53
The Operations Management literature has also identified some major problems with the accuracy of the inventory
records themselves, which will result in suboptimal inventory management decisions. DeHoratius and Raman (2008)
find that retail firms only have an accurate inventory record for about 35% of their products and that inventory records

62
firms, and result in low performance on inventory related decisions. SOX Section 404 mandates

that US listed firms report material weaknesses in internal control. The absence of a clear policy

with respect to the access to the warehouse or problems with the methodology of order-picking

could lead to an inventory-related material weakness in internal control. Feng et al. (2015b) show

that firms with inventory-related material internal control weaknesses have lower inventory

turnover ratios and are more likely to report inventory impairments. They also find that firms that

remediate these inventory-related material weaknesses report increases in sales, gross profit and

operating cash flows.

7.3. Competition management decisions: desirable properties of costing information

Gal-Or (1993) introduces the idea that a firm’s competitive environment may drive firms

to strategically allocate costs to favor products sold in markets where strategic considerations are

of relatively greater importance. If a firm sells one product in an oligopolistic market and another

product in a perfectly competitive market, it may benefit from under allocating cost to the product

sold in the oligopolistic market and shifting those costs to the product sold in the perfectly

competitive market. This deliberate misallocation will permit the firm to strategically commit to

an aggressive production plan that makes its rival cut its own production in the oligopolistic

market, whereas the firm is a powerless price taker in the competitive market anyway.54 Banker

and Potter (1993) think about a similar issue in terms of choosing between a more accurate ABC

system or a single cost driver system that is inaccurate. They show that for a monopolist it is

important to have the accurate information, while they argue that firms would prefer a costing

are larger than physical inventory nearly as often as they are smaller. While most inventory management models in
the Operations Management literature assume that the inventory records are accurate, Mersereau (2013) develops a
replenishment model based on inaccurate inventory records.
54
Gal-Or (1993) models how the level of competition in the oligopolistic market affects the firm’s choice for under
or over allocating costs.

63
system that inflates all costs in an oligopolistic market, as it would allow these firms to circumvent

the preclusion of explicit collusion by anti-trust rules.55 However, since costing is a zero sum game

whereby some products will be over costed while others are under costed, they require a simulation

method to suggest the conditions under which increased overall error may be beneficial in

oligopolistic markets.

Callahan and Gabriel (1998) distinguish between Cournot and Bertrand competition,

arguing that the former firms are playing a cost leadership strategy while the latter follow a

differentiation strategy. They show that the former benefit from improved cost information that is

more accurate, while the latter do not. However, Hansen (1998) shows that in a Cournot market,

a firm’s demand for improved cost data is not monotonically increasing in competition, but

displays a U-shape instead. The initial decline of the demand for better cost information stems

from the issue that as each firm’s sales drop, the cost reduction investment can only be recovered

on fewer units. I refer the reader to the Foundations and Trends in Accounting monograph by

Narayanan and Smith (2013) for a complete development on this topic.

Experimental researchers have also studied the desirability of cost information accuracy

for the management of competition, particularly focusing on following a specific competitive

environment over time and incorporating the idea that seeing how the market plays out provides

informational feedback. For example, King and Gupta (1997) focus on a monopolist, while

Buchheit (2004) experimentally models a duopoly. Krishnan et al. (2002) find that firms’

convergence to a particular level of cost accuracy depends on the market structure and the degree

of competition. Cardinaels et al. (2004) include the notion that firms learn from competitors with

superior information, while Cardinaels et al. (2008) show that it is beneficial for the leader to have

55
Hughes and Kao (1998) also model how firms use cost allocation to coordinate tacitly.

64
accurate cost information, but that this is not the case for the follower. Lastly, the analytical

modeling literature has studied the role for full costing in procurement auctions (e.g., Cohen and

Loeb 1990; Budde and Göx 1999).

8. Identification of cost management opportunities

8.1. Measurement object and which resource costs to include for identification of cost

management opportunities

The cost management function aims to maintain effective financial control of projects and

processes. The use of standard costs that are carefully predetermined costs and set as the budgeted

costs that should be obtained by diligently performing the firms’ production processes. is

recommended here. Already in the eighties, 86% of manufacturing companies were reported to

use standard cost (Cornick et al. 1988). Standard costs are used as the building block in flexible

budgeting, which, in contrast to a static budget, adjusts or “flexes” for changes in the volume of

activity by calculating the costs that would have been allowed had an accurate prediction about

volume been made at the start of the budgeting process. A flexible budgeting analysis then presents

the cost information in a format amenable to a variance analysis, where we can pinpoint the sources

of variation between actual and budgeted performance (e.g. price, fixed cost, variable cost or

efficiency variation), assign responsibility, and take remediating action.56 Unfortunately, while

management accounting textbooks advocate flexible budgeting and variance analysis, we have no

recent survey evidence available to know whether firms have adopted these practices widely or

not.57

56
I will discuss the aspect of assigning responsibility when using variance analysis more elaborately in Section 10.2.2.
57
The evidence on the prevalent use of standard costs may suggests that many firms are in the position to potentially
perform variance analyses. Simons (1987) asks a survey question on the use of cost variances in cost control, but does
not report descriptive statistics. Some (e.g. Cheatham and Cheatham 1996) have argued that using variance analysis

65
 Early literature on cost variance investigation indeed mostly concerned controlling of the

production process. Kaplan (1975) presents an overview of the early models put forward for such

cost variance investigation. Hughes (1975) adds the optimal timing of the cost report as an

additional feature to consider in variance investigation, while Brown (1981) studies the manager’s

information processing of the reported variances. Jacobs (1978) experimentally evaluates various

variance investigation models. Others devise various improvements to the technique such as

production function efficiencies (Marcinko and Petri 1984), the role of substitution (e.g. for

cheaper inputs) in the production function (Darrough 1988), the optimal level of disaggregation of

the cost variances (Livnat et al. 1980), and the rules of thumb that drive which variances to further

investigate (Chow et al. 1990). Current applications of flexible budgets and variance analysis

frequently follow the production processes and assign responsibilities for dealing with any

deviations from the plan at the appropriate level. I will discuss the aspect of assigning

responsibility when using variance analysis more elaborately in Section 10.2.2 of the chapter on

performance measurement. For further explanation on how flexible budgeting works, and some

thoughts on how these standard costs can be updated dynamically, I refer the reader to Cheatham

and Cheatham (1996).

At the operational level, the collection of non-financial indicators of performance such as

defects and quality issues typically complements such analysis, and we separate out non-

controllable costs. Usually, standard costs only include factory and manufacturing costs, and not

resources consumed to design, market and deliver the product (Kaplan 1988). Therefore, in more

recent years and with the advent of ABC, cost management activities have been broadened up to

is mutually exclusive with using ABC, limiting its adoption. However, Kaplan (1994) shows how both costing
techniques can be integrated. See also Mak and Roush (1994) and Kloock and Schiller (1997).

66
also include such non-manufacturing resources and consider cost management in the longer run,

even if such analysis may not happen as frequently as the day-to-day operational control of the

production processes.

When used for such more strategic and long run cost management, the focus of the ABC

analysis normally shifts from the cost object level to the resource level, as the ultimate objective

is to reduce the overall resource costs in the firm. For long run cost management purposes, the

distinction that ABC makes between cost driver use and cost rate for each activity is very useful.

ABC expresses total cost TC as 𝑇𝐶 = ∑K∈B ∑J∈tu 𝑟J × 𝑑J , whereby

H set of ABC hierarchical levels, index l

Al set of activities performed on ABC hierarchical level l, index a

ra cost rate per activity a expressed in $ for one unit of activity a

da driver use expressed in units of activity a performed

As an example of an activity at order level, consider the delivery activity. We can reduce driver

use at this activity, for example, by combining more customer deliveries in one truck-tip making

use of scheduling algorithms or by imposing a minimum order quantity on customers. We can

reduce the cost rate per delivery activity by replacing an old delivery fleet with more fuel efficient

trucks or by negotiating reduced fringe benefits with the truck drivers. A thorough cost

management exercise will go through each activity at each hierarchical level and consider ways to

both reduce cost rate and cost driver use for each activity. Some activities may be altogether

considered non-value adding to the firm, and hence be eliminated completely, which is equivalent

with bringing the cost driver use to zero. In following this exhaustive Activity-Based Cost

Management framework, no stone will be left unturned in the identification of potential

opportunities for cost reduction. Again, Bob Kaplan took the lead in extending ABC to Activity-

67
Based Management, and I refer the reader to some early case examples described in Cooper et al.

(1992).

The “strategic cost management” literature (Shank 1989) adds to this perspective by

introducing the value chain concept in an attempt to move management accounting somewhat

away from its internal focus to also look externally for cost management opportunities, in

particular from the perspective of what does or does not create value for the customer. Customers

are only prepared to pay for product or service attributes to which they attach value, and hence any

cost management exercise should start with eliminating activities that are not value creating

(McNair et al. 2001).58 Shank (1989) also looks at the other end of the value chain to cost

management opportunities that can be explored by working with suppliers. This aspect generated

a lot of interest in accounting research, and I will cover interorganizational cost management

separately in Section 8.3.

Further, the strategic cost management perspective also brings into focus how firm strategy

affects the measurement object choice of the costing system. For example, a firm following a cost

leadership strategy in a mature industry will need to maintain good cost control via a constant

comparing of standard costs and actual cost and do thorough competitor cost analyses. However,

a firm that follows a product differentiation strategy has limited need for the calculation of standard

costs and techniques like flexible budgeting, whereas this firm will need to analyze marketing

costs in depth to keep its costs at reasonable levels.59 For Shank’s deeper development of the

strategic cost management and value chain perspective, see also Shank and Govindarajan (1992).

58
However, it may be impossible to eliminate all activities that do not create customer value. For example, a firm is
required to prepare a tax return and has to support this activity, even though it is not of value to the customer.
59
The strategic cost management literature also introduces the terminology of structural cost drivers versus executional
cost drivers. The first type help identify cost management opportunities consistent with firm strategy, while the latter
type are tools to implement and incentivize the cost management strategies. This distinction goes back to the core

68
 8.2. Desirable properties of costing information for identification of cost management

opportunities

As suggested by the ABC cost management discussion above, the focus of the costing

system for cost management purposes is less on the accuracy of the reported product costs than on

the process view of the activities performed in the firm that drive resource consumption and

ultimately generate resource costs. Since activities and processes change frequently, both because

of exogenous factors such as changes in prices in input commodity markets as well as endogenous

changes because of cost management actions taken in the firm, frequent reporting and updating of

this cost information will be necessary to support this objective. I refer the reader back to Section

4.2 where I call for research to make further progress on understanding costing system updating,

which is extremely relevant for the cost management purpose.

Kaizen cost management, a Japanese cost management technique exported to the West,

embodies this continuous improvement perspective on cost management, by striving for cost

reductions throughout all processes. The Kaizen cost management technique is focused on three

different sources of cost reduction options, continuously (Yoshikawa and Kouhy 2012). First, a

Just in Time approach reduces not only inventory costs, but also prevents the holding of stock to

hide underlying organizational problems. Second, resource consumption is expressed in physical

terms, allowing for an immediate focus on resource consumption reductions. Third, reducing waste

from performing non-value added activities, owning excess capacity, and bad management

practices that soak up resources unnecessarily are central to its implementation. I refer the reader

to case examples of Kaizen cost management explored in Yoshikawa and Kouhy (2012) and the

issue of presenting cost information both for decision-making and performance measurement, albeit specifically with
a focus on decisions and incentives with relation to cost management.

69
references therein. Unfortunately, there is limited academic research on Kaizen cost management

practices, and we do not know how prevalent the technique is. Furthermore, existing plant-floor

costing practices such as ABC tend to consider each workstation independently, which makes it

impossible to measure the cost of forced idleness of one workstation because of a disruption at

another workstation appropriately. To resolve this issue, Bai et al. (2016) develop a cost

measurement method based on real-time throughput analysis.60

TDC also is an advantageous technique for cost management purposes because it embraces

the process view of the firm at its core since it starts with drawing workflow maps of the activities

performed in the firm. In particular, such workflow maps help identify the variation in these

activities and what drives that variation, so that it can be captured in the time equations that form

the basis of the cost calculations (Balakrishnan et al. 2017). For example, first time patients take

much longer when being seen by a physician in a clinic than repeat patients (Scott et al. 2018).

Understanding such variation is a first step towards managing processes in a more cost effective

manner. Scott et al. (2018) report on the clinic’s introduction of a frequently asked questions email

send to first time patients prior to their visit in order to reduce the time needed to go through routine

issues with the physician. The case study by Hoozée and Bruggeman (2010) demonstrates the

importance of early employee involvement in the TDC development to reap the cost management

benefits by identification of operational improvements.

8.3. Interorganizational cost management

The value chain perspective brought into focus by strategic cost management proponents

such as Shank (1989) and Shank and Govindarajan (1992) naturally led to the inclusion of the

60
For another example of the effect of production interdependencies on cost drivers, see Leitch (2001).

70
supplier-end of the value chain in the identification of cost management opportunities, and a large

literature on interorganizational cost management was born. Questions addressed in this literature

include the impact of various supply chain decisions on cost management, such as the make or buy

choice, the degree of vertical integration of the supply chain, the selection of suppliers, the type of

partnership entered into with the supplier, the extent to which accounting records are transparent

in the supply chain, and how reverse logistics are handled (Anderson and Dekker 2009). Cooper

and Slagmulder (2004) describe how various relational contexts allow supply chain partners to use

different interorganizational cost management practices, which are focused on improving cost

management decisions’ effectiveness. On the other hand, Baiman and Rajan (2002) concentrate

on the performance measurement and incentive effects of supply chain and product architecture

choices, also covering their effect on cost reductions.

Tools advocated for interorganizational cost management are Vendor Managed Inventory

(VMI), concurrent cost management, and Total Cost of Ownership (TCO). Under VMI, the

supplier takes control of the buyer’s inventory management practices, necessitating the exchange

of detailed information on inventory levels and production demand forecasts (Kulp 2002).

Concurrent cost management entails the involvement of the supplier’s engineers early in the design

of the buyer’s product and will be further discussed in the next Section (9.4) on cost management

in product design. TCO considers the entire cost of working with a particular supplier, including

delivery costs, quality costs, stockout risk costs and the like, moving beyond simply price paid (de

Boer et al. 2001; Degraeve et al. 2004). Indeed, often costs associated with supply chain

disruptions or quality issues outweigh price reductions obtained in negotiations with suppliers.

TCO offers an example of a cost management tool that serves both cost management decision-

making purposes as well as performance measurement purposes. The quantification of the total

71
cost incurred with the selection of suppliers supports not only supplier selection decisions

(Degraeve et al. 2005) and decisions on where to focus joint cost reduction efforts, but also

negotiations with suppliers (Van den Abbeele et al. 2009; Masschelein et al. 2012), and

performance evaluation of suppliers (Wouters et al. 2005). Interorganizational cost information

sharing is a recurring theme in this literature (e.g. Fayard et al. 2012).

This Section only presents a taste of the voluminous literature on interorganizational cost

management. For a systematic overview and further references, I refer to Anderson and Dekker

(2009) and Jakobsen (2012).

8.4. Cost management in the design of products and target costing

A smaller literature also developed on cost management methods used during product

development. By far the most popular method is target costing (Wouters and Morales 2014), which

is a Management Accounting costing method that specifies the allowed cost for a new to be

developed product with specific functionality and quality while ensuring the desired profit margin.

Unlike cost-based pricing which starts from the newly developed product cost and adds the

required margin to determine sales price, target costing is market-driven and starts with market

research to understand the functionalities and specifications the customers desire and the price they

are willing to pay for these. Thus, the technique fits very well with the perspective of strategic cost

management proponents. If, at first, the target cost seems unachievable, product designers,

engineers, and management accountants go through repeated value engineering processes,

functional cost analysis and other cost reduction strategies to eventually, if successful, reach the

72
cost goal and have the opportunity to launch the new product. Achieving this cost goal depends

critically on the collaborative effort of all parties mentioned.61

Management accounting, product development and supply chain management publications

have all suggested that the largest proportion of costs are committed or set in stone during the

product design phase, while the actual cash outflows only happen much later when buying

materials, equipment, and labor (Cooper and Kaplan 1998a). Conventional wisdom suggests an

80/20 rule with 80% of costs committed during the first 20% of a product’s lifecycle, roughly

coinciding with the design and development phase (Cooper and Chew 1996).62 It is then no surprise

that both management accounting and supply chain management experts have advocated the early

involvement of management accountants and suppliers in product design, respectively (e.g.,

Anderson and Sedatole 1998; Wynstra et al. 1999). For a review of the literature on early supplier

involvement in new product development, I refer the reader to Johnsen (2009).

Tools such as target costing and life cycle costing have been developed to support such

endeavors (Cooper and Kaplan 1998a; Kajuter 2012; Monden and Hamada 1991). Life cycle

costing goes a step further than target costing in an attempt to estimate costs over the entire life

cycle of the product, including the final salvaging phase, bringing the temporal dimension of cost

management into play. The core problematic issue, even with management accountants involved

early in the design phase, however, remains the measurement object of interest. Since the products

and services that are being designed are not produced yet, the management accountant is faced

with having to predict costs based on very little data. The long-term orientation introduced by life

61
This section draws from Dierynck and Labro (2018).
62
However, Labro (2006b) suggests that there is actually no empirical evidence to back up this claim, while the only
existing empirical evidence suggests a much more muted importance of the design phase in cost commitment of 50%
(Ulrich and Pearson 1998).

73
cycle costing only adds to the likelihood that there will be substantial estimation error in this

process. Surprisingly, literature on how to improve the accuracy of the initial forward-looking

estimate is lacking, while most literature instead focuses on cost management methods to close the

gap between that cost estimate and the price that the market is prepared to pay for a product or

service with a particular set of attributes.

Various case studies describe target costing applications (e.g., Patell 1987; Davila and

Wouters 2004; Woods et al. 2012). Although most case studies cover target costing in a

manufacturing setting, Askarany and Yazdifar (2012)’s survey of 584 CIMA members finds that

target costing is equally prevalent in the service sector in the UK, Australia and New Zealand, with

an overall full adoption rate of about 18%, with an additional 6% adopting on a trial basis.

Gopalakrishnan et al. (2015) experimentally find that target cost goal specificity helps obtain larger

cost reductions in a sequential new product development process; specific goals do not aid in that

same way in a concurrent process. Mihm (2010) studies which incentive schemes can best combine

with target costing to achieve new product development objectives.

Navissi and Sridharan (2017) provide a recent overview of the target costing literature,

while Wouters and Morales (2014) and Wouters et al. (2016) provide extensive overviews of the

broader literature on all cost management methods used during product development.

9. Inventory valuation for financial and tax accounting

9.1. Measurement object and which resource costs to include for inventory valuation

Inventory valuation for financial and tax accounting requires the allocation of periodic

production costs between goods sold, goods in inventory, and work-in-progress, following

74
generally accepted accounting principles and rules. These rules do not require a causality in such

allocation and the use of a single cost allocation base for the entire plant (Kaplan 1988). Auditors

will simply focus on a reasonable split of the overhead factory costs between cost of goods sold

and goods and work-in-progress still owned by the plant (Kaplan 1988). The resources allocated

(factory costs) for the purpose of inventory valuation are more limited than the resources allocated

for other purposes discussed previously.

9.2. Desirable properties of costing information for inventory valuation

While all the prior objectives discussed require a costing approach that is focused on

explaining the cause of resource consumption, inventory valuation for financial and tax accounting

objectives simply requires that all overhead costs are allocated. In fact, since cost accounting’s

original objective related mostly to the provision of cost numbers to be used for inventory

valuation, Miller and Vollmann (1985) blame the disproportional growth of overhead costs on the

fact that the cost models used for inventory valuation simply allocate costs rather than explain

costs.

Desirable properties of reported product costs are also different from the characteristics of

good cost information discussed in the context of other objectives discussed earlier. More

aggregate cost allocations are sufficient, and there is no need for item-level cost information

(Kaplan 1988). Timeliness is less important, with yearly or quarterly reporting frequencies being

adequate. Furthermore, frequent updating of the costing system is unnecessary.

75
10. Performance measurement and control demand for costing systems

10.1. Measurement object and which resource costs to include for performance

measurement and control

Once decisions are taken that will maximize firm value, the firm’s owners need to ensure

that all employees of the firm are aligned behind these decisions and will implement them.

Therefore, employees’ performance towards obtaining these goals needs to be measured, and

control actions taken in case of deviating behavior. For obtaining an understanding of the control

and performance measurement use of information in general, and cost information specifically,

the literature typically resorts to agency theory. I will present a stripped down version of an agency

model to provide the reader who is unfamiliar with this theory with the basic ideas and tradeoffs.63

In this very basic agency model, a Principal with utility function G contracts with a risk averse

Agent with utility function U to exert unobservable effort in return for a wage which is a function

of the observable outcome 𝑥$ . A self-interested Agent will only behave in the way desired by the

Principal when his pay structure is set up in such way that Principal and Agent’s incentives are

aligned. A discrete outcome (N different 𝑥$ values) – binary input measure (𝑎8 for high effort, 𝑎K

for low effort) version of the classic agency model by Holmström (1979), is specified as follows:

𝑀𝑎𝑥A]( ∑y
$./ 𝐺 (𝑥$ − 𝑆$ )𝑝(𝑥$ |𝑎8 )

s.t.

∑y
$./ 𝑈(𝑆$ )𝑝(𝑥$ |𝑎8 ) − 𝑉(𝑎8 ) ≥ 𝑈 (IR)

∑y y
$./ 𝑈(𝑆$ )𝑝(𝑥$ |𝑎8 ) − 𝑉(𝑎8 ) ≥ ∑$./ 𝑈(𝑆$ ))𝑝(𝑥$ |𝑎K ) − 𝑉 (𝑎K ) (IC)

63
The remainder of this paragraph is drawn from Hemmer and Labro (2017).

76
V(a) is the cost of effort of the Agent and p is the conditional probability of achieving a particular

outcome, given the level of effort exerted. The Principal’s (say, CEO or shareholders of the firm)

objective is to maximize his residual, while offering the Agent (say, a worker in the firm) a wage

that satisfies both his individual rationality (IR) constraint which ensures that he earns at least his

reservation utility 𝑈 and his incentive compatibility (IC) constraint which makes the Agent prefer

to put in high effort rather than low effort.64 The key is that the contract imposes some risk on the

Agent because it is written on an observable performance measure that is not necessarily revealing

the amount of effort the Agent put in fully accurately.

The outcome x in this agency model is the performance measure that will be used to

incentivize the Agent to put in high effort, or, in an extended model, to choose the desired tasks to

perform.65 The exact nature of the performance measure x is not as important, and could be any

form of financial or non-financial information. In fact, Antle and Demski (1988) show that the

choice to measure performance based on cost, revenue or profit reports is not determined by the

type of measure or the resources allocated by the costing system but by how much the Principal

can learn from observing the measure about the effort level of the Agent. In lieu of the

controllability principle that says that an employee should only be held accountable for things

under his control, Antle and Demski (1988) propose to apply Holmström (1979)’s informativeness

principle to this accounting setting. Even performance measures that are not directly impacted by

the Agent can still be useful in contracting, as long as they are conditionally controllable by the

Agent. In such case, the information content of the measure is controllable, given that we condition

64
Note that in this simple version of the agency model, the problem is set up such that the Principal always prefers
the Agent to put in high effort because the benefits of the output generated by high effort outweigh the cost of
incentivizing the Agent for high effort.
65
Note that, technically, the Principal doesn’t learn the Agent’s effort level from observing the performance measure
score, but that the incentive compatibility (IC) constraint that is written on the performance measure that will later
become observable ensures the Agent will indeed put in high effort.

77
on the other information that is available, and hence the measure will be informative about the

Agent’s action choice.

Having said that, some literature specifically relates to the cost information that can be

used in performance measurement and control. In parallel with the literature that developed on

explanations for the use of full costing (rather than marginal costing) in decision-making contexts,

a literature developed to explain full costing and other cost allocations observed in practice to

support performance measurement and control. Zimmerman (1979)’s dissertation shows that such

cost allocations can serve to control various agency problems, such as managers’ overconsumption

of the firm’s resources as perquisites. Furthermore, he shows how the allocation of a superior’s

expenses to subordinates incentivizes the subordinates to monitor the superior’s expenditures and

reduce over-utilization of a common resource in a decentralized organization.66

Baiman and Noel (1985) apply the above concept of non-controllable costs to capacity

costs in a multi-period model where the Principal makes all investment/disinvestment decisions

related to capacity, leaving the Agent only with control over operating responsibilities. They show

the conditions under which it is optimal to base the Agent’s subsequent compensation on the

current realized non-controllable capacity costs. Suh (1987) models a firm with two sequential

divisions, where the cost of the intermediate product division are non-controllable by the final

product division. In his setting, there is an opportunity for collusion between the two divisions,

where the (costly) quality of the intermediate product can be set higher than the Principal desires.

In order to discourage such collusion, it can be optimal for the Principal to hold the manager of

66
While not immediately related to cost allocation, Kanodia and Dickhaut (1989) show that it is rational for a manager
who is trying to hide private information on his human capital from the labor market to escalate an already made
investment (a sunk cost) rather than abandoning it and switching to an investment that would be more value enhancing
for the firm. Switching out would reveal that the manager is of the low talent type.

78
the final product division responsible for the non-controllable intermediate costs. While the prior

work supports the use of cost allocation in an agency setting, it remains devoid of practical

guidance on how to calculate these allocations. Linking allocations to the methods described in

management accounting textbooks, Hemmer (1996) finds that the optimal capacity cost allocation

is only a function of budgeted volume when capacity can be used to produce only a single product.

However, in a joint production setting, the optimal allocation is based on the joint products’

estimated net realizable values.

10.2. Desirable properties of costing information for performance measurement and

control

Returning to agency theory, it is clear that the desirable property of costing information is

no longer accuracy, yet informativeness about the Agent’s action choice. Better performance

measures allow the Principal to contract with the Agent at lower cost, as less risk needs to be

imposed on the risk averse agent, and hence a lower risk premium can be paid. Hence, the quality

of the performance measure depends on how the likelihoods that high effort versus low effort led

to a certain performance measure score compare. I formalize this notion by forming the Lagrangian

of the optimization problem presented in Section 10.1., and by taking the first order condition

(FOC) with respect to 𝑆$ (Labro 2015). I obtain

𝑑𝐿
=0⇔
𝑑𝑆$

−𝐺 } (𝑥$ − 𝑆$ )𝑝(𝑥$ |𝑎8 ) + 𝜆𝑈 } (𝑆$ )𝑝(𝑥$ |𝑎8 ) + 𝜇𝑈 } (𝑆$ )€𝑝(𝑥$ |𝑎8 ) − 𝑝(𝑥$ |𝑎K )• = 0

This expression further simplifies to

𝑑𝐿 𝐺 } (𝑥$ − 𝑆$ ) 𝑝(𝑥$ |𝑎8 ) − 𝑝(𝑥$ |𝑎K )

= 0 ⇔ = 𝜆 + 𝜇
𝑑𝑆$ 𝑈 } (𝑆$ ) 𝑝(𝑥$ |𝑎8 )

79
 8 K
1€𝑥$ ‚𝑎 •ƒ1€𝑥$ ‚𝑎 •
In this expression, the term 8 is the likelihood ratio, which indicates the precision
1€𝑥$ ‚𝑎 •

with which the outcome 𝑥$ indicates that the effort level exerted was 𝑎8 . Viewing 𝑥$ as a

performance measure of the Agent’s effort, the precision of the performance measure determines

its quality and hence impacts the weight placed on the measure in the contract 𝑆$ . All else equal, a

measure with higher precision is preferred since it reduces the risk that is imposed on the (typically

considered) risk averse Agent (Hemmer and Labro 2017).

10.2.1. Multiple performance measures

In practice, multiple performance measures are typically available. Holmström (1979)

shows that any publically observed piece of information that does not satisfy the sufficient statistic

condition generates value for an agency.67 When y is publically observable, using both x and y

(instead of x alone) in the performance evaluation will lead to a Pareto improvement, if and only

if it is not the case that x is a sufficient statistic for y. Mathematically, the sufficient statistic

condition holds when 𝑝(𝑥, 𝑦|𝑎) = 𝑝(𝑦|𝑥) × 𝑝(𝑥|𝑎). In this case, we learn nothing new from

observing the additional performance measure y that we do not already learn from observing x,

and hence y will not be used. To see this, substitute the sufficient statistic condition in the

likelihood ratio:

𝑝(𝑦% |𝑥$ ) 𝑝(𝑥$ |𝑎8 ) − 𝑝(𝑦% |𝑥$ )𝑝(𝑥$ |𝑎K ) 𝑝(𝑥$ |𝑎8 ) − 𝑝(𝑥$ |𝑎K )
=
𝑝(𝑦% |𝑥$ )𝑝(𝑥$ |𝑎8 ) 𝑝(𝑥$ |𝑎8 )

Holmström (2017) explains that such informativeness of an additional measure y can arise

through two channels. Let 𝑧̃ = 𝑥‡ + 𝜖̃ denote the firm’s output where (as before) the distribution

of x‡ is determined by the binary effort level 𝑎 and ϵ‡ is pure noise. Then, a signal y‡ is useful for

67
This paragraph is drawn from Hemmer and Labro (2017).

80
contracting either if its distribution is informative about x
Œ, and thereby about the part of firm

performance that is determined by 𝑎, but also if it is informative about 𝜖̃. Several measures of

reported cost can hence typically be useful, as many cost measures that can be calculated do not

satisfy the sufficient statistic condition.68 Note, though, that in an agency setting the value of

information can be positive and neutral, but also negative, even at zero cost to produce such

information. Given the multi-person setting, we cannot simply ignore information, and additional

information can on occasion make contracting costlier. For a series of examples demonstrating

these scenarios, see Christensen (1982). Mishra and Vaysman (2001) illustrates a setting in which

a traditional (less accurate) costing system is preferred to an ABC system driven by the

informational rents that need to be paid to the Agent under the ABC system.

Two aspects of the performance (or cost) measure determine its quality and hence its

desirability. First, ceteris paribus, a performance measure for which 𝑝(𝑥$ |𝑎8 ) is high increases the

likelihood that the particular desired score on the performance measure 𝑥$ is obtained by an Agent

exerting high effort, strengthening the link between high effort and a good score, and reducing the

risk imposed on the Agent. Second, again ceteris paribus, a performance measure for which

𝑝(𝑥$ |𝑎K ) is low decreases the likelihood that an Agent who exerts low effort can generate the same

desired score on the performance metric. Hence, it is less likely that the Principal will confuse a

shirking lazy Agent with a hard working Agent, again reducing the risk for the Agent who puts in

high effort.

68
In a model where the Agent not only choses unobservable effort but also an observable level of utilization of a
resource supplied by the Principal, Magee (1988) shows that the optimal compensation function must include the
resource level as an argument, and hence include some form of cost allocation. That is, the Agent will be compensated
based on 𝑥 − 𝑐(𝑦), where x is output generated and c(y) is the allocated cost.

81
 10.2.2. Variance analysis and flexible budgeting

The performance measurement objective of costing makes heavy use of standard costs that

are carefully predetermined costs that are set as the budgeted costs that should be obtained by

diligently performing the firms’ production processes, as introduced in Section 10.1. While the

early literature on cost variance investigation mostly concerned controlling of the production

process in a single decision-maker context, the concept underlying these variance investigation

models is that an Agent is responsible for the variance(s) and needs to take action. Lambert (1985)

is the first to explicitly introduce variance investigation in an agency model, while Magee and

Dickhaut (1978) run an experiment that puts the practice in the context of different compensation

schemes. Other papers that have linked cost variances with performance appraisals and contracting

include Gietzmann and Selby (1994), Budde (2009) and Hansen (2011). Empirical work on

flexible budgeting and variance analysis is limited, with some notable exceptions provided by

Gribbin and Lau (1991), Bai (2016) and Davila and Wouters (2005).

10.3. Deliberate misallocation to circumvent performance measurement and control

Any time cost information is used in performance measurement and control, the risk exists

that those cost numbers will be manipulated to make the person or firm being evaluated look like

a better performer than they really are. There exists a vast theoretical and empirical literature on

such manipulation behavior, but in what follows I will provide some examples of papers that have

explicitly researched how performance measure manipulators use cost reports and cost allocations.

Many of these empirical studies are done in the health care sector, where data availability

and changing regulations allow researchers to pinpoint cost shifting issues. Cost allocation may

determine hospitals’ revenues (which are reimbursements by insurance companies, the

government or patients). For example, a hospital that serves two patient populations, poor elderly

82
patients (paid a fixed amount per case by Medicare) and affluent maternity cases (reimbursed by

private insurers based on cost) will choose, as much as possible, allocation bases that allocate more

costs to the maternity patients. In the US before 1983 reimbursements by private medical insurance

companies and government (Medicare and Medicaid) were based on costs. In 1983, Medicare

changed this cost-based reimbursement to a fixed payment per Diagnosis-Related Group (DRG)

for inpatient services, while outpatient services continued to be paid on reported costs. Eldenburg

and Kallapur (1997) document, as a consequence, both real changes in patient treatment, where

outpatient services such as lab tests and X-rays are unbundled from those provided during inpatient

stays so that these services can continue to benefit from cost-based reimbursement, and overhead

allocation changes. Soon after 1983, private insurers moved away from cost-based reimbursement

too.

Krishnan and Yetman (2011) study California non-profit hospitals who have an incentive

to report relatively higher program expenses compared to fundraising and administrative expenses

in order to look “lean” to potential donors. They find that substantial cost shifting is done on the

IRS 990 form that serves as the source data for charity websites, as compared to the activity reports

these hospitals have to file with the Office of Statewide Health Planning and Development, with a

ratio of program to total expenses of 86.5% and 77.4%, respectively.69 Predictably, hospitals that

rely more on donations as a percentage of revenue shift costs more. The program for Critical

Access Hospitals (CAH), which supports small rural hospitals to guarantee access to care, provides

a contemporaneous example of the strategic changes in cost allocations to manipulate

reimbursements. Medicare reimburses CAHs based on costs. Ederhof (2014) finds that when a

69
Program expenses are costs related to providing the nonprofit organization's programs or services in accordance
with its defined mission.

83
hospital system gets to convert one of their rural hospitals to a CAH, “home office costs” (an

allocation of support resources provided by the hospital system to that hospital) jump by 12%. Her

calculations suggest that there is a total increase of administrative and general costs for all CAH

that are part of a hospital system of ca. $150 million, which represents a substantial 5% of Medicare

payments to CAHs in these hospital systems.70

Such cost shifting behavior also exists in other sectors. Using a large dataset from the

National Center of Charitable Statistics, Jones and Roberts (2006) find that charitable

organizations use joint cost allocations to manage their program ratio.71 At a time where defense

contracts were still mostly paid by the US government on a cost-plus basis, they would have an

incentive to allocate more costs to the government contract, if they also worked on private contracts

(Demski and Magee 1992). Lanen et al. (2008, pp. 216) find evidence in line with this prediction.

A private shipyard that did work on both US Navy ships on a cost-plus basis and private fixed-fee

contracts used the age of its dry dock facilities to base their decision on where to do the work.

Newer facilities that have much higher depreciation and overhead rates were chosen for

government projects. On the other hand, McGowan and Vendrzyk (2002) are unable to find

evidence that the abnormally high profits of defense contractors during 1984-1989 were caused by

shifting costs to cost-plus contracts with the Department of Defense (DoD). Federal Reserve Banks

are an agency of the US government and as such required by law to charge private banks fees for

their services based on direct and allocated costs. Cavalluzzo et al. (1998) find that the Fed shifted

allocation of costs from competitive services (e.g. check-clearing) to less competitive services (e.g.

electronic transfers). These empirical results are generally consistent with the theories described

70
Other references that provide evidence of cost shifting in the health care sector include Hwang and Kirby (1994),
Eldenburg et al. (2011), and Hsu and Qu (2012).
71
See also Krishnan et al. (2006). Cost allocation rules can also be the topic of renegotiation attempts in universities
(Modell 2006).

84
in Section 7.3 on how competition may affect cost allocation. Other examples are the deliberate

overestimation by employees of the expense budget needed to be able to fulfill their job-related

duties so as to build in slack in their budget and the deliberate overcosting of unique components

to incentivize the use of common components by product designers (Merchant and Shields 1993;

Alles et al. 1998; Cooper and Kaplan 1987). Experimental research has also frequently used a cost

reporting setting to study participants’ incentives and honesty in reporting (Evans III et al. 2001).

As Eldenburg and Kallapur (1997), discussed above, suggest, sometimes firms take real

actions to affect reported costs in a particular direction, rather than just misreport the cost numbers.

The classical example here is overproduction in firms with high manufacturing overhead in order

to delay expensing overhead into earnings by capitalizing it into inventory (Gupta et al. 2010). It

is notoriously difficult to separate optimal decision-making in the face of uncertainty (e.g. a

delayed reaction or inability to respond to negative demand shocks) from opportunistic decision-

making, and it is hence not surprising that some studies suggest that overproduction is an optimal

reaction (Jiambalvo et al. 1997) whereas others argue it is suboptimal and evidence of

opportunistic behavior (e.g., Roychowdhury 2006; Gupta et al. 2010).

85
PART C: INTERACTIONS AND CONFLICTS BETWEEN SOURCES OF DEMAND FOR

COST INFORMATION AND IMPLICATIONS FOR COSTING SYSTEM DESIGN

11. Interactions and conflicts between sources of demand for cost information

11.1. Conflicts

Table 2 summarizes the measurement object, the resource cost considered and the desirable

properties of the costing system as demanded by the different objectives for which cost information

can be used, as discussed in Part B. Note that these desirable properties are only clearly desirable

if they are costless to supply. If not, each specific demand-side objective will need to be understood

to weigh off the benefits and the cost of such properties. However, it is easy to see from this

summary table that there will be conflicting requirements for the support of different objectives,

even if these properties are costless!

Table 2: Summary on measurement object, resource cost and desirable properties of costing
systems
Objective Measurement object Resource cost Desirable properties
Decision-making Choice variable of Marginal cost only Accuracy
decision at cost
Timeliness
object level

Cost management Cost level of all All Process view of

resources available activities
and consumed Frequent updating
Inventory valuation Goods sold, Factory cost Aggregate
inventory, and work-
Less timely
in-progress cost
Performance Agent’s unobservable Role for full cost Informativeness
measurement action choice allocations

Almost a century ago, Clark (1923) already pondered the difficulty in providing

harmonized cost information that achieves each objective and concluded with proposing “different

86
costs for different purposes”. Developing multiple costing systems geared at the ideal properties

for each purpose seems indeed to be a potential solution to the issue of the conflicts previously

described. However, developing and maintaining multiple costing systems is expensive (Al-Omiri

and Drury 2007), so financially constrained firms may not want to go down the path of delivering

different types of cost information for different objectives. Furthermore, even without financial

constraints, it is possible that firms prefer to work with one costing system so as to not confuse

their employees who use the cost information by presenting different costs for the same cost

object,72 create extra work to justify and reconcile differences, or attract the attention of the tax

authorities.73 Lastly, humans are boundedly rational, and face difficulties processing lots of cost

information and accomplishing conflicting objectives, so firm management may opt to focus on

the most important objective(s) for the firm only, and tailor the properties of the costing system to

that ultimate objective. Hence, costing system choices will reveal management’s preferences and

the relative importance they attach to the different objectives costing systems serve.

Empirical evidence seems to be in line with firms indeed not pursuing a strategy of

developing multiple costing systems. Johnson and Kaplan (1987) point out that in general only

one costing system is used in the US and blame the dominance of the inventory valuation objective

in the design of that one system for low quality decision-making and cost management efforts. In

the United Kingdom, Brierley et al. (2001) conclude that firms also mostly use just one costing

system, but that this system may not necessarily be serving the inventory valuation objective. For

example, Drury and Tayles (2006)’s UK survey suggest that product profitability calculations to

support decision-making are the costing system’s prime objective, and only 12% of companies

72
For example, Hays PLC’s CFO is quoted as saying “We all have one version of the truth” in a recent Wall Street
Journal article (Trentmann and Minaya 2018).
73
I will develop this latter point further in Chapter 13 on transfer pricing.

87
were unable to calculate product profitability at individual product level. Granlund and Lukka

(1998) and Brierley et al. (2001) point out that with the rise of ERP systems, it has become easier

to report cost information generated by the same underlying system in different ways, thereby

adapting its properties towards each objective.

11.2. Interactions

While the prior Section suggests a conflict between the demands on cost information for

different purposes, different demands may also simply interact in a neutral way, and interactions

may potentially be beneficial. Most of the research that studies neutral or beneficial interactions

like these has been at the higher-level construct of internal information quality, going beyond just

cost information quality. While in the next three Chapters I will illustrate those interactions that

are specific to cost information, it is useful to think about the higher-level concept of internal

information quality and how different objectives interact with it as this may generate further

research ideas that are specific to costing.

Most of this research is on the interface between managerial and financial accounting, as

that is where financial information objectives are distinctly different. Financial accounting is

mostly concerned with investors’ ability to value the firm accurately, with the shareholders’

understanding of how well the CEO and his team provide stewardship of the firm, and with

supporting decision-making by external (to the firm) stakeholders such as (potential) investors,

regulators, suppliers and possibly even customers. Managerial accounting, on the other hand, is

concerned with the ability of managers and employees internal to the firm to make good decisions

and measure performance on firm goals. However, empirically, there is a tight link between

managerial and financial accounting, as reported by 80% of managers surveyed by Dichev et al.

(2013).

88
 Some research has documented that changes in the financial accounting objectives affect

managerial accounting information quality. Hemmer and Labro (2008) develop a model that links

the objectives of financial accounting, and the choices the accounting regulator makes on that front

between stewardship and valuation focus, to the properties of the optimal managerial accounting

system that solely supports managerial decision-making to create firm value, while performance

measurement of the manager is only based on the financial accounting system.74 They show that

if the financial accounting regulator’s objective is sufficiently slanted towards valuation focus, the

optimal accuracy of the managerial accounting system will become very high, while when instead

the regulator’s objective is more slanted towards a stewardship focus, management accounting

quality will be lower. Empirically, Shroff (2017) finds that some changes in GAAP alter managers’

information sets and changes their investment decisions, in particular their capital and R&D

expenditures. Likewise, Cheng et al. (2018) find that the adoption of SFAS 142 on goodwill

induces firms to collect more information in order to be able to comply with the regulation, which

in turn improves the internal information environment as evidenced by improved merger and

acquisition quality, internal capital allocation efficiency, and performance. Feng et al. (2015b) find

that ineffective internal control over financial reporting, as required to be disclosed by the

Sarbanes-Oxley (SOX) regulations of 2003, affects firm performance negatively. In particular,

their results suggest that firms with inventory-related material weaknesses in internal control over

financial reporting have weak inventory management practices, in that they have lower inventory

turnover ratios and are more likely to report inventory impairments. Wagner and Dittmar (2006)

document how the increased stewardship value information brought on by SOX results in

companies developing better information systems to support their operations and avoid making

74
This is frequently the case with CEO compensation linked contractually to financial accounting numbers such as
earnings.

89
bad decisions. Cho (2015) finds that diversified firms that change their segment definitions upon

adoption of SFAS 131, which required firms to define segments as internally viewed by managers,

experienced an improvement in internal capital allocation efficiency as the more transparent

segment information helps resolve internal agency conflicts.

Other research has studied the opposite directional effect, where changes in managerial

accounting quality affect financial accounting quality.75 Cassar and Gibson (2008) find that small

privately-held firms make more accurate revenue forecasts if they have an internal accounting

reporting and budgeting process in place. Dorantes et al. (2013) show that firms that implemented

an improvement to their internal information systems increased their forecast accuracy. Chen et

al. (2018) find that internal information asymmetry within conglomerate firms lowers the quality

(accuracy, bias, specificity and frequency) of management forecasts and increases the likelihood

of error-driven financial accounting restatements. Ittner and Michels (2017) find that firms with

more sophisticated risk-based forecasting and planning processes have lower earnings forecast

errors and narrower forecast widths. Lastly, Gallemore and Labro (2015) link the quality of the

internal information environment in with another use of accounting information: tax planning.

They find that firms with high quality internal information environments can do better tax planning

because they are more likely to be able to coordinate across divisions, deal with uncertainty and

document supporting evidence to provide to the tax authorities. Indeed, such firms have lower and

less variable effective tax rates.

75
Hemmer and Labro (2019, forthcoming) show analytically that firms making managerial decisions based on internal
information that subsequently gets reported externally in aggregate form less frequently affects the behavior of
earnings response coefficients and the earnings frequency distribution, and can explain several financial accounting
puzzles.

90
 Moving back to a focus on costing systems specifically, in the next four chapters, I will

develop on specific interactions and conflicts between demands for costing systems. Chapter 12

will develop the conflict between decision-making and performance evaluation further, while

Chapter 13 expands on the multiple objectives faced with transfer pricing and the role of costs

therein. Lastly, Chapter 14 focuses on inventory, and the three conflicting objectives centered on

inventory: valuation, cost control, and decision-making. For each chapter, I will first lay out the

particular issue and conflicting requirements for costing system characteristics together with

research documenting such conflict, after which I will cover research that attempts to integrate the

specific purposes. I selected the topics of these specific chapters to illustrate conflicts and

interactions based on the prevalence of existing literature and its depth, but they are by no means

an exhaustive list of the conflicting and interacting demands on cost information that can be

fruitfully researched.76

12. Interactions and conflicts between demands on cost information for performance

evaluation and decision-making

12.1. Conflicts

The most far-reaching conflict exists between the two general purposes of decision-making

and performance evaluation. Comparing the general implications for the use of cost information

presented in Chapter 5 (decision-making) and Chapter 10 (performance measurement) makes clear

that the requirements are fundamentally different. For decision-making, observing a finer piece

76
As an example of a study that links the cost management and incentive objectives, using a laboratory experiment,
Drake et al. (1999) document that the benefits of providing ABC-information to realize cost reductions are dependent
on the incentive system that the employees are subject to, and whether or not that incentivize cooperation or
competition.

91
cost information y will allow the decision maker to recompute his subjective probability

distribution over the state space, which is driven by the difference between the probability p(x)

and the conditional probability p(x|y). Any information that allows him to improve predictive

power about the future state of the world is valuable, and costless information never has negative

value, as it can simply be ignored. For performance measurement, informativeness of an additional

measure y can arise through two channels. a signal y‡ is useful for contracting either if its

distribution is informative about x

Œ, and thereby about the part of firm performance that is

determined by 𝑎, but also if it is informative about 𝜖̃. It is hence important for this purpose what

the Principal learns about the effort of the Agent, rather than about the state of the world. Even

costless information can have negative value because, given the multi-person context, it cannot

simply be ignored. I refer the reader to Chen et al. (2010) for a fully developed model on the

interaction between both purposes within the context of the firm, also making predictions on how

this interaction endogenously creates bias in accounting information.

The prime management accounting example where there is a clear conflict between both

purposes is in budgeting, which serves both a decision-making or planning purpose and a

performance measurement purpose, and calculates costs to support both objectives.77 An

omnipresent practice in companies, it is surrounded with controversy, partly because the difficulty

of achieving both purposes using one budget. One the one hand, for decision-making and planning

purposes, the budget aims to predict the resources necessary (or superfluous) to produce the

planned product mix and quantities in the next period. For this objective, it is important to obtain

accurate estimates of the resource consumption patterns and unbiased predictions of demand. On

77
Ewert and Wagenhofer (2014) talk about this as a separate objective for cost information under the heading
“coordination”.

92
the other hand, this budget subsequently becomes the standard against which to evaluate

performance of employees in achieving those plans within cost constraints. Using the budget for

performance measurement objectives, however, will result in employees overestimating the

resources needed in the future and underestimating the amount of sales they can make so that they

can build in slack in the performance standard against which they will be evaluated. Such biased

cost data and predictions are detrimental to good decision-making and planning.

Therefore, most firms tailor their budgeting system a little more towards one of the two

objectives, depending on where they face the most pressing need for cost information. Firms

focusing on its use in planning are more likely to use a bottom-up approach whereby cost, demand

and resource consumption information is learned from employees most closely related to the

production of the product or service. These sources have more accurate information than

management which is further removed from the day-to-day operations have. However, with such

bottom-up approach, the budget cannot become the standard against which to measure employee

performance, or else employees will sandbag. Firms focusing on the use of budgets in performance

measurement and incentives, on the other hand, will prefer to use a top down approach where

management sets budgets without the input of the employees. In this case, while the advantage is

that there is no biased input, the disadvantage is that the quality of cost information generated for

decision-making is lower. For a more elaborate analysis of choices on how to design budgeting

systems under different conflicting objectives, see Hansen (2011).

12.2. Interactions

By far the largest literature that connects decision-making and incentives through cost

allocation choices is the theoretical literature on the optimization of capacity planning and

allocation decisions in settings where multiple divisions have conflicting incentives. The basic

93
setting is one where divisions are able to more accurately estimate demand, and hence resource

needs, than headquarters. In such setting, deliberate biases in cost information may arise

endogenously. In a single period model under excess capacity, it is optimal for headquarters to

only charge variable costs to ensure utilization of the excess capacity by divisions as the

opportunity cost of that idle capacity equals its variable cost. This creates a deliberate undercosting

bias, since no fixed costs are allocated to the division. However, looking at a model with multiple

periods where in the first period headquarters need to decide how much capacity to acquire based

on the inputs by divisions on their demand predictions and resource needs and, in the second

period, divisions utilize the procured capacity, changes inferences. Charging only variable costs in

the second period when utilization of the previously acquired capacity is incentivized will give

divisions the incentive to overestimate demand and resource needs in the first period with the aim

to create excess capacity for which they subsequently then only pay variable cost. To manage this

issue, headquarters will allocate (part of) the fixed costs in the first period to discipline managers

from seeking excess capacity. In sum, there is a tradeoff between efficient investment and

utilization of capacity: depreciation (or the allocation of fixed costs) controls overinvestment in

the first period but creates capacity under-utilization in the second period. The optimal bias in

costing, then, depends on the importance of the over-investment versus the under-utilization

problem. Allocating fixed costs make sense if over-investment is the bigger problem, while only

using variable costs is optimal when excess capacity exists and under-utilization is most important

(Dierynck and Labro 2018).

This literature goes on to show what preferences are for accounting measurement in such

context, and establishes roles for measures of residual income, alternative depreciation choices,

various cost allocation schemes, and historical cost numbers. While the above describes the basic

94
results on costs allocations in multi-divisional firms facing capacity planning and allocation

decisions and misaligned incentives in dynamic settings, this literature has been developed in much

more detail (e.g., Rajan 1992; Pfaff 1994; Anctil et al. 1998; Wei 2004; Baldenius et al. 2007;

Dutta and Reichelstein 2010; Göx 2010; Ray and Goldmanis 2012; Rajan and Reichelstein 2009).

I refer the reader to the Foundations & Trends in Accounting monographs by Dutta (2007) and

Bastian Johnson and Pfeiffer (2016) which provide extensive coverage of this literature.

13. Transfer Pricing: Interactions and conflicts between tax reporting and managerial

accounting demands on cost information

13.1. Conflicts

13.1.1. Management accounting focus

The managerial accounting literature on transfer pricing has focused on its use in creating

optimal production incentives in multi-divisional companies, aligning head quarters’ objectives

with divisional managers’ objectives. This literature was jumpstarted by the seminal article by

Hirshleifer (1956) who models a decentralized firm with a headquarter and two divisions. The

upstream seller produces an intermediate product and supplies it to the downstream buyer division

that processes and sells in the final product market. Both divisions are organized as profit centers,

and will hence maximize their own profits rather than those of the firm. The aim of headquarters

is to set a transfer price that coordinates the decisions of divisions so that it maximizes aggregate

firm profit. Hirshleifer (1956) shows that this can be done by setting the transfer price equal to the

opportunity cost of the intermediate product. The question then becomes what this opportunity

cost is. If there is a competitive market for the internal product, the opportunity cost is the market

95
price, while, if no market price exists, the optimal transfer price is the marginal cost of the

intermediate product.

Of course, the same difficulty as developed on in Chapter 6 obtains in calculating that

marginal cost. Holmström and Tirole (1991) aptly point it out: “The economist’s first instinct is to

set the transfer price equal to marginal cost. But it may be difficult to find out marginal cost. As a

practical matter, marginal cost information is rarely known to anybody in the firm, because it

depends on opportunity costs that vary with capacity use. And even if marginal cost information

were available, there is no guarantee that it would be revealed in a truthful fashion for the purpose

of determining an optimal transfer price.” As a consequence, also here management accounting

researchers looked into the role of marginal cost approximations observed in practice such as full

costing, variable costing, standard costs, actual cost, etc. I refer the reader to section 4 in Göx and

Schiller (2007) for an overview of the analytical literature on this topic.78

As Göx and Schiller (2007) develop on more deeply, Hirshleifer (1956) does not contain

an economic rationale for having a divisionalized firm in the first place. With a perfectly

competitive market, independent firms could achieve the same profit. Without a perfectly

competitive market, a fully integrated firm could achieve the same profit without needing to resort

to setting transfer prices to solve the coordination problem, as headquarters could simply dictate

quantities to be bought and sold. Göx and Schiller (2007) provide an overview of how management

accounting researchers hence started focusing on frictions such as information asymmetry (or even

simply information processing costs) and adverse selection to introduce an advantage of

78
Papers include, among others, Cohen and Loeb (1988), Pfeiffer et al. (2011), Baldenius et al. (1999), Sahay (2003),
Dikolli and Vaysman (2006) and Vaysman (1996).

96
decentralization over centralization. These models also can explain why cost mark-ups above

marginal costs are used, as they grant informational rents to the better-informed party.

Incomplete contracting models which start from the premise that it is impossible to foresee

every single contingency can explain why negotiated transfer prices are used.79 Because the

allocation of decision rights can shift the bargaining power such that it prevents a hold-up effect

and underinvestment, these type models also support a role for allocating decision authority.

However, this is not a fully satisfactory answer to why decentralized organizational structures are

observed as the same agreements can be set up in a situation where two independent firms trade

(Göx and Schiller 2007). Another answer lies in the strategic use of transfer pricing as a

commitment device in competition (Alles and Datar 1998; Göx 2000). For certain types of market

competition, it is possible to set transfer prices above marginal costs to reduce the intensity of

competition in the final product market and obtain profits that could otherwise only be obtained

through collusion between competitors. Such strategic transfer pricing models can again explain

the prevalence of mark-ups on marginal cost and the use of full cost based transfer prices.80 While

this offers a rationale for having a decentralized firm, this reasoning requires that the competitor(s)

observe the internal transfer price, which is an unlikely assumption. For an exhaustive overview

of the management accounting literature on transfer pricing, I refer the reader to the Foundations

and Trends in Accounting piece by Baldenius (2009).

13.1.2. Tax Accounting focus

Tax accounting research has typically focused on the use of transfer prices as a mechanism

for multi-national companies to shift costs to divisions in high tax regimes and away from divisions

79
Arya et al. (2017) study the role of joint cost allocation in such setting.
80
On the other hand, Matsui (2013) shows that computing transfer prices based on an observable direct costing system,
and hence committing to low transfer prices, can deter competitive entry.

97
in low tax regimes to minimize taxes paid for the overall firm. Particularly when there is no

competitive market for the intermediate good, firms may have some leeway in documenting an

arms-length transaction price, and may use cost allocations strategically to show more income in

a low tax jurisdiction, while showing less income in a high tax jurisdiction. For example, Bernard

et al. (2006) document that U.S-based multinationals set prices substantially larger for arm’s length

customers than for internal related-party trade, and that this price wedge is larger when goods are

send to countries with lower corporate tax rates, consistent with the strategic determination of

transfer prices to minimize global tax paid. This mostly empirical literature is vast, and nicely

summarized in the Foundations and Trends in Accounting piece by Blouin (2011), chapter 4.

13.1.3. One Set of Books or Two?

Unfortunately, the management accounting literature and the tax accounting literature on

transfer pricing seem to have developed mostly separately, focusing on their respective core

interests in production incentives and tax minimization, with both literatures considering the use

of transfer prices for the other objective to be second order, either implicitly or explicitly.81

However, research has analytically shown that the same set of transfer prices is unlikely to

optimize both objectives jointly (Reichelstein and Hiemann 2012; Baldenius et al. 2004; Göx and

Schiller 2007).82 Often researchers have argued that a firm can simply use two sets of books and

hence different transfer prices to optimize both objectives and circumvent this problem.

81
I will cover some exemplar exceptions to this statement in the next section.
82
An additional hurdle in bridging the managerial accounting literature and the tax literature on transfer pricing is
methodological: the vast majority of the former are analytical studies while the largest proportion of the latter are
empirical studies, mostly using archival methods, complicating the connection between the two literatures. There are,
of course, some exceptions on both sides. For example some management accounting research on the topic is empirical
and uses case (Cools and Slagmulder 2009; Rossing and Rohde 2010) or survey methods (Chen et al. 2015; Bouwens
and Steens 2016), while some tax research is analytical (Sansing 1999).

98
 Practice, however, seems to suggest otherwise and indicates that the use of a single transfer

price is most common (Herzig 2012; Cools and Slagmulder 2009; Reichelstein and Hiemann 2012,

and references therein). As before, it may simply be too costly to maintain two sets of books, with

two different cost allocation systems. It is also possible that there is a lack of internal acceptance

of the use of dual transfer prices (Göx and Schiller 2007). Furthermore, current anecdotal evidence

suggests that such decoupling of the books is a red flag triggering the attention of the tax

authorities, who may ask to access the internal books and use differences in the cost numbers used

internally with those used in the tax filings as evidence that the income reported in the tax filings

is not based on arms-length transfer prices, imposing increased tax risk (Herzig 2012). Rossing

and Rohde (2010) present evidence that adopting a transfer pricing tax compliance strategy led

their case study firm to make substantial changes to its costing system. The firm defined cost

objects at business unit level rather than divisional level, clarified the definitions of cost pools,

increased the number of allocation bases used, and increased the frequency of cost reporting. The

firm also engaged in an Advanced Pricing Agreement with the tax authority to ex ante approve of

the cost allocation bases it wanted to use.

13.2. Interactions

Calls for a closer connection of the tax and managerial accounting literatures on transfer

pricing have been made in prominent overview pieces (Shackelford and Shevlin 2001; Göx and

Schiller 2007; Blouin 2011). A recent Ernst & Young survey on transfer pricing (Griffin et al.

2017) indicates that more than one third of practitioners, too, perceive the need for greater

integration between the tax and business aspects of transfer pricing. In contrast to the simplifying

assumption that both objectives (managerial and tax), and hence the demands for cost information,

can simply be decoupled, the tax admissible transfer price is in itself part of the economically

99
relevant valuation of the transaction in question. Tax reporting for a particular transaction results

in cash flows to tax authorities and after tax cash flows determine the overall value of the

transaction (Hiemann and Reichelstein 2012). Therefore, identifying a suitable internal transfer

price needs to take account of the firm’s transfer price used for tax reporting purposes, and vice

versa (Hiemann and Reichelstein 2012).

Baldenius et al. (2004) show that when one single transfer price attempts to balance the

conflicting goals of tax minimization and efficient resource allocation among divisions, the

optimal transfer price is a weighted average of the pre-tax unit cost and the most favorable arm’s

length price that can be justified to the tax authorities. The difficulty in identifying this optimal

transfer price stems from the effect of the transfer price on quantities being transferred. While the

most favorable arm’s length price minimizes the firm’s overall tax liability for given transfer

quantities ex ante, intra-company transfers will be too low whenever this arm’s length price

exceeds the pre-tax unit cost. As a result, expected corporate profit after taxes may be maximized

by a transfer price that does not minimize the firm’s tax liability ex post. Chapter 5 in the

Foundations and Trends piece by Sansing (2014) provides an overview of the analytical modeling

literature on this topic.

Section 13.1.1. briefly described the gist of the theoretical management accounting

literature on transfer pricing and concluded with the thought that this literature has struggled to

include an economic rationale for having divisionalized firms compete in the first place. The most

promising way so far to resolve this issue, in my opinion, lies in bridging this literature with the

tax use of transfer prices. The literature has shown that transfer prices can be strategically used as

commitment devices to soften competition and increase the surplus for competing firms (Alles and

Datar 1998; Göx 2000). However, without the internal transfer price being observable by all

100
competing parties, commitment to a particular transfer price is not credible. Narayanan and Smith

(2000) show that the use of a specific transfer price for taxation reasons can deliver such credible

commitment. Operating in different tax jurisdictions provides a credible signal that the transfer

price set will be larger than the marginal cost of production, providing a very reasonable way to

explain a joint choice for decentralization and transfer pricing. More research on this interface will

provide a fruitful path forward to the transfer pricing literature.83

14. Conflicts and interactions between demands for cost information for inventory

valuation, management and control

14.1. Conflicts

From the elaboration in Section 7.2 on inventory management decisions and in Chapter 10

on inventory valuation, it is immediately clear that there is a large conflict between the demands

of cost information from both objectives. On the one hand, optimizing inventory management

decisions requires a detailed and disaggregate view on all costs, including procurement costs,

preferably via the use of an ABC hierarchy, and ideally even the possibility of exchanging this

information with vendors (as discussed in Section 9.3). On the other hand, for inventory valuation,

the allocation of factory costs at a very aggregate level suffices, and limited requirements on the

causality and accuracy of the allocation bases used are imposed. Furthermore, timeliness and

frequency of reporting is of utmost importance for inventory management, while low periodicity

of reports is sufficient for inventory valuation. Conflict additionally arises because the inventory

83
More generally, albeit not specifically related to costing systems, I believe the interface between managerial
accounting and tax is a fruitful one, as already demonstrated by several studies, e.g. Robinson et al. (2010), Phillips
(2003), and Gallemore and Labro (2015).

101
valuation objectives affect real inventory management decisions. For example, Gupta et al. (2010)

show that firms overproduce inventory to delay expensing overhead into earnings by capitalizing

it into inventory, thereby showing the firm to be more profitable.

An additional conflict arises when we consider the objective of inventory control, and the

incentives that arise for agents working with the inventory. Chen and Sandino (2012) document

that stealing among employees is quite prevalent and can be partially explained by variation in

wage levels among employees, in that employees who feel unfairly paid will resort to stealing.

DeHoratius and Raman (2007) document that decreasing the weight on inventory shrinkage in the

compensation plan of store managers leads to an increase in inventory shrinkage. Hence, inventory

levels are endogenously determined by other organizational design choices such as compensation

practices. In order to deal with such inventory control problems, timely inventory cost reporting is

necessary, and the source of this report ought to be unrelated to the agents who deal with the

inventory on a day-to-day basis and may have opportunities and incentives to steal. Good

inventory-related internal control processes are a solution here. For example, the absence of a clear

policy with respect to the access to the warehouse or problems with the methodology of order-

picking could lead to a public US firm having to report an inventory-related material weakness

under SOX Section 404. The use of technology such as RFID in the inventory recording process

is also useful, as it will help control the inventory reporting bias that may be introduced through

human intervention in reporting. For example, an employee stealing inventory will have an

incentive to upwardly bias inventory records to conceal the theft.

14.2 Interactions

There exists some research showing a positive interaction between the inventory control

and inventory management objectives. Feng et al. (2015b) show that firms with inventory-related

102
material weaknesses in internal controls have lower inventory turnover ratios and are more likely

to report inventory impairments. They also find that firms that remediate these inventory-related

material weaknesses report increases in sales, gross profit and operating cash flows, improving

their inventory-related decision-making through the increased inventory information quality. This

research, however, does not dig down to the level of the preferred costing system attributes, and it

would be interesting to get more insights into the specific role of costing systems in supporting

multiple inventory-related objectives.

15. Conclusion

Costing practice shows a lot of variety, including some that seems suboptimal at first sight.

This monograph studies the properties of costing systems from the perspective of the source of the

demand for costing information. Different demands for cost information impact the preferred

measurement object of the costing system, which resources to include in the cost measurement,

and the desired properties of the costing system. For decision-making, the measurement object is

the choice variable of the decision at cost object level, and only marginal costs ought to be included

in the calculations. The most voluminous literature here is on the capacity acquisition and

allocation decision problem, and a substantial part of this literature is devoted to explaining how

this theoretical demand for marginal cost only calculations can be compatible with full costing

practice. Other decisions featured in academic research on the demands on costing systems are

customer portfolio decisions, inventory management decisions and decisions on managing

competition. Desirable properties for costing systems for decision making mostly relate to

accuracy, granularity and timeliness.

103
 For cost management purposes, we move away from cost objects as the measurement

object, yet focus on the cost level of the resources that are available for use and that are consumed,

and also attempt to measure the cost of the unused resources. Typically, firms should analyze all

resource costs and frequently update their costing systems, which preferably provide a process

view of activities. Next, for inventory valuation for financial accounting and tax purposes, the

measurement object is limited to understanding the cost of goods sold, of inventory and of work-

in-progress, with a parallel limited focus on solely factory costs. Less timely and more aggregate

cost reports are sufficient. For performance measurement purposes, the desired measurement

object is the Agent’s unobservable action choice, and literature has established a rationale for the

role of cost allocations in providing information about this unobservable feature.

Firms often have resource and behavioral constraints on using multiple costing systems.

When one costing system is used to support multiple objectives, it is clear that conflicting demands

are placed on costing systems’ measurement objects, resource costs considered and desired costing

system properties. Hence, firms will make choices that trade off the importance of each objective

for their costing system, under the financial constraints imposed, leading to a variety of costing

practices observed. Hence, observing different costing choices may reveal firms’ preference to

more strongly support one objective over another. The monograph also carefully points out

research that has demonstrated that sometimes these objectives do not necessarily conflict, yet

merely interact, possibly even in a positive way, demonstrating avenues for further interesting

research on costing systems.

104
 16. References

Agthe, K. 1959. Stufenweise Fixkostendeckung im System des Direct Costing. Zeitschrift für
Betriebswirtschaft (July):742-748.
Al-Omiri, M., and C. Drury. 2007. A survey of factors influencing the choice of product costing
systems in UK organizations. Management Accounting Research 18 (4):399-424.
Al-Qady, M., and S. El-Helbawy. 2016. Integrating Target Costing and Resource Consumption
Accounting. Journal of Applied Management Accounting Research 14 (1):39-54.
Al-Sayed, M., and D. Dugdale. 2016. Activity-Based Innovations in the UK Manufacturing
Sector: Extent, Adoption Process Patterns and Contingency Factors. The British
Accounting Review 48 (1):38-58.
Alles, M., S. Datar, and M. Gupta. 1998. Tradeoffs Within Costing Systems Between Incentives
and Measurement Objectives. Managerial Finance 24 (1):1-18.
Alles, M., and S. M. Datar. 1998. Strategic Transfer Pricing. Management Science 44 (4):451-
461.
Anand, V., R. Balakrishnan, and E. Labro. 2014. Updating Cost Systems. In Working Paper.
Anand, V., R. Balakrishnan, and E. Labro. 2017. Obtaining Informationally Consistent Decisions
When Computing Costs with Limited Information. Production and Operations
Management 26 (2):211-230.
———. 2018, forthcoming. A Framework for Conducting Numerical Experiments on Cost
System Design. Journal of Management Accounting Research:46 pp.
Anctil, R. M., J. S. Jordan, and A. Mukherji. 1998. Activity-Based COsting for Economic Value
Added. Review of Accounting Studies 2:231-264.
Anderson, S. W. 1995. Measuring the Impact of Product Mix Heterogeneity on Manufacturing
Overhead Cost. The Accounting Review 70 (3):363-387.
Anderson, S. W., and H. C. Dekker. 2009. Strategic Cost Management in Supply Chains, Part 1:
Structural Cost Management. Accounting Horizons 23 (2):201-220.
Anderson, S. W., and K. L. Sedatole. 1998. Designing Quality into Products: The Use of
Accounting Data in New Product Development. Accounting Horizons 12 (3):213-233.
———. 2013. Evidence on the Cost Hierarchy: The Association between Resource
Consumption and Production Activities. Journal of Managment Accounting Research 25
(1):119–141.
Antle, R., and J. S. Demski. 1988. The Controllability Principle in Reponsibility Accounting.
The Accounting Review 63 (4):700-718.
Anupindi, R., and L. Jiang. 2008. Capacity Investment Under Postponement Strategies, Market
Competition, and Demand Uncertainty. Management Science 54 (11):1876-1890.
Arya, A., J. C. Glover, and B. Mittendorf. 2017. The Effects of Joint Cost Allocation on Intra-
Firm Trade: A Comparison of Insulating and Non-Insulating Approaches. Journal of
Management Accounting Research 29 (2):1-10.
Askarany, D., and H. Yazdifar. 2012. An Investigation into the Mixed Reported Adoption Rates
for ABC: Evidence from Australia, New Zealand and the UK. International Journal of
Production Economics 135 (1):430-439.
Atkinson, A. A. 1987. Discussion of "On The Allocation of Fixed and Variable Costs From
Service Departments". Contemporary Accounting Research 4 (1):186-193.
Babad, Y. M., and B. V. Balachandran. 1993. Cost Driver Optimization in Activity-Based
Costing. The Accounting Review 68 (3):563-575.

105
Bai, G. 2016. Applying Variance Analysis to Understand California Hospitals' Expense
Recovery Status by Patient Groups. Accounting Horizons 30 (2):211-223.
Bai, G., T. Kajiwara, and J. Liu. 2016. Measuring the Cost of Individual Disruptions in
Multistage Manufacturing Systems. Journal of Management Accounting Research 28
(1):1-26.
Baiman, S., and J. Noel. 1985. Noncontrollable Costs and Responsibility Accounting. Journal of
Accounting Research 23 (2):486-501.
Baiman, S., and M. V. Rajan. 2002. Incentive Issues in Inter-firm relationships. Accounting,
Organizations and Society 27:213-238.
Baker, K. R., M. J. Magazine, and H. L. W. Nuttle. 1986. The Effect of Commonality on Safety
Stock in a Simple Inventory Model. Management Science 32 (8):982-988.
Balachandran, B. V., R. Balakrishnan, and K. Sivaramakrishnan. 1997. On the Efficiency of
Cost-Based Decision Rules for Capacity Planning. The Accounting Review 72 (4):599-
619.
Balachandran, B. V., L. Li, and R. P. Magee. 1987. On the Allocation of Fixed and Variable
Costs From Service Departments. Contemporary Accounting Research 4 (1):164-185.
Balachandran, B. V., and R. T. S. Ramakrishnan. 1981. Joint Cost Allocation: A Unified
Approach. The Accounting Review 56 (1):85-96.
———. 1996. Joint Cost Allocation for Multiple Lots. Management Science 42 (2):247-258.
Balakrishnan, R., and D. V. DeJong. 1993. The Role of Cost Allocations in the Aquisition and
Use of Common Resources. Contemporary Accounting Research 9 (2):395-414.
Balakrishnan, R., T. S. Gruca, and D. Nath. 1996. The Effect of Service Capability on Operating
Costs: An Empirical Study of Ontario Hospitals. Contemporary Accounting Research 13
(1):177-207.
Balakrishnan, R., S. Hansen, and E. Labro. 2011. Evaluating Heuristics Used When Designing
Product Costing Systems. Management Science 57 (3):520-541.
Balakrishnan, R., D. M. Koehler, and A. S. Shah. 2018. TDABC: Lessons from an Application
in Healthcare. Accounting Horizons forthcoming.
Balakrishnan, R., E. Labro, and K. Sivaramakrishnan. 2012a. Product Costs as Decision Aids:
An Analysis of Alternative Approaches (Part 1). Accounting Horizons 26 (1):1-20.
———. 2012b. Product Costs as Decision Aids: An Analysis of Alternative Approaches (Part 2).
Accounting Horizons 26 (1):21-41.
Balakrishnan, R., A. J. Pugely, and A. S. Shah. 2017. Modeling Resource Use with Time
Equations: Empirical Evidence. Journal of Management Accounting Research 29 (1):1-
12.
Balakrishnan, R., and K. Sivaramakrishnan. 1996. Is Assigning Capacity Costs to Individual
Products Necessary for Capacity Planning? Accounting Horizons 10 (3):1-11.
Balakrishnan, R., and K. Sivaramakrishnan. 2001. Sequential solutions to capacity planning and
pricing decisions. Contemporary Accounting Research 18 (1):1-25.
Balakrishnan, R., and K. Sivaramakrishnan. 2002. A critical overview of the use of full-cost data
for planning and pricing. Journal of Management Accounting Research 14:2-31.
Balakrishnan, R., K. Sivaramakrishnan, and G. Sprinkle. 2008. Managerial Accounting. 2nd ed:
Wiley.
Baldenius, T. 2009. Internal Pricing. Foundations and Trends in Accounting 3 (4):223-313.
Baldenius, T., S. Dutta, and S. Reichelstein. 2007. Cost Allocation for Capital Budgeting
Decisions. The Accounting Review 82 (4):837-867.

106
Baldenius, T., N. Melumad, and S. Reichelstein. 2004. Integrating Managerial and Tax
Objectives in Transfer Pricing. The Accounting Review 79 (3):591-615.
Baldenius, T., S. Reichelstein, and S. A. Sahay. 1999. Negotiated versus Cost-Based Transfer
Pricing. Review of Accounting Studies 4:67-91.
Banker, R., and S. Hansen. 2002. The Adequacy of Full-Cost-Based Pricing Heuristics. Journal
of Management Accounting Research 14:34-58.
Banker, R., and J. Hughes. 1994. Product costing and pricing. The Accounting Review 69
(July):479-494.
Banker, R., I. Hwang, and B. K. Mishra. 2002. Product Costing and Pricing under Long-Term
Capacity Commitment. Journal of Management Accounting Research 14:79-97.
Banker, R., G. Potter, and R. G. Schroeder. 1995. An Empirical Analysis of Manufacturing
Overhead Cost Drivers. Journal of Accounting and Economics 19:115-137.
Banker, R. D., and H. H. Johnston. 1993. An Empirical Study of Cost Drivers in the U.S. Airline
Industry The Accounting Review 68 (3):576-601.
Banker, R. D., and G. Potter. 1993. Economic Implications of Single Cost Driver Systems.
Journal of Management Accounting Research 5 (Fall):15-24.
Bastian Johnson, N., and T. Pfeiffer. 2016. Capital Budgeting and Divisional Performance
Measurement. Foundations and Trends in Accounting 10 (1):1-100.
Bernard, A. B., J. B. Jensen, and P. K. Schott. 2006. Transfer Pricing by U.S.-Based
Multinational Firms. In NBER working paper, 36.
Biddle, G. C., and R. Steinberg. 1984. Allocations of Joint and Commom Costs. Journal of
Accounting Literature 3.
Billera, L. J., D. C. Heath, and R. E. Verrecchia. 1981. A unique procedure for allocating
common costs from a production process. . Journal of Accounting Research 19 (1):185-
196.
Bjornenak, T. 1997. Diffusion and Accounting : the Case of ABC in Norway. Management
Accounting Research 8:3-17.
Bjornerak, T., and O. Helgesen. 2012. Customer Relations and Cost Management. In The
Routledge Companion to Cost Management, edited by J. Morten and M. Falconer: Taylor
& Francis, 250-265.
Blouin, J. 2011. Taxation of Multinational Corporations. Foundations and Trends in Accounting
6 (1):1-64.
Bonacchi, M., K. Kolev, and B. Lev. 2015. Customer Franchise-A Hidden, Yet Crucial, Asset.
Contemporary Accounting Research 32 (3):1024-1049.
Borle, S., S. S. Singh, and D. C. Jain. 2008. Customer Lifetime Value Measurement.
Management Science 54 (1):100-112.
Bouwens, J., and B. Steens. 2016. Full-Cost Transfer Pricing and Cost Management. Journal of
Management Accounting Research 28 (3):63-81.
Brierley, J. A. 2008. Toward an Understanding of the Sophistication of Product Costing Systems.
Journal of Management Accounting Research 20:61-78.
Brierley, J. A., C. J. Cowton, and C. Drury. 2001. Research Into Product Costing Practice: A
European Perspective. The European Accounting Review 10 (2):215-256.
Bromwich, M., and C. Hong. 1999. Activity-based costing systems and incremental costs.
Management Accounting Research 10 (1):39-60.
Brown, C. 1981. Human Information Processing for Decisions to Investigate Cost Variances.
Journal of Accounting Research 19 (1):62-85.

107
Brüggen, A., and J. L. Luft. 2016. Cost Estimates, Cost Overruns, and Project Continuation
Decisions. The Accounting Review 91 (3):793-810.
Bruns, W. J., and R. S. Kaplan. 1987. Accounting & Management: Field Study Perspectives.
Boston: Harvard Business School.
Buchheit, S. 2004. Fixed Cost Magnitude, Fixed Cost Reporting Format, and Competitive
Pricing Decisions: Some Experimental Evidence. Contemporary Accounting Research 21
(1):2-24.
Budde, J. 2009. Variance analysis and linear contracts in agencies with distorted performance
measures. Management Accounting Research 20 (3):166-176.
Budde, J., and R. F. Göx. 1999. The Impact of Capacity Costs on Bidding Strategies in
Procurement Auctions. Review of Accounting Studies 4:5-13.
Cagwin, D., and M. J. Bouwman. 2002. The Association Between Activity-Based Costing and
Improvement in Financial Performance. Management Accounting Research 13 (1):1-39.
Callahan, C. M., and E. A. Gabriel 1998. The Differential Impact of Accurate Product Cost
Information in Imperfectly Competitibve Markets: A Theoretical and Empirical
Investigation. Contemporary Accounting Research 15 (4):419-453.
Campbell, D., and F. Frei. 2010. Cost Structure, Customer Profitability, and Retention
Implications of Self-Service Distribution Channels: Evidence from Customer Behavior in
an Online Banking Channel. Management Science 56 (1):4-24.
Cardinaels, E. 2008. The interplay between cost accounting knowledge and presentation formats
in cost-based decision-making. Accounting, Organizations and Society 33 (6):582-602.
Cardinaels, E., and E. Labro. 2008. On the Determinants of Measurement Error in Time-Driven
Costing. The Accounting Review 83 (3):735-756.
Cardinaels, E., F. Roodhooft, and L. Warlop. 2004. The Value of Activity-Based Costing in
Competitve Pricing Decisions. Journal of Management Accounting Research 16:113-148.
Cardinaels, E., F. Roodhooft, L. Warlop, and G. Van Herck. 2008. Competitive Pricing in
Markets with Different Overhead Costs: Concealment or Leakage of Cost Information?
Journal of Accounting Research 46 (4):761-784.
Casas-Arce, P., F. A. Martínez-Jerez, and V. G. Narayanan. 2017. The Impact of Forward-
Looking Metrics on Employee Decision-Making: The Case of Customer Lifetime Value.
The Accounting Review 92 (3):31-56.
Cassar, G., and B. Gibson. 2008. Budgets, Internal Reports, and Manager Forecast Accuracay.
Contemporary Accounting Research 25 (3):707-737.
Cavalluzzo, K. S., C. D. Ittner, and D. F. Larcker. 1998. Competition, Efficiency, and Cost
Allocation in Government Agencies: Evidence on the Federal Reserve System Journal of
Accounting Research 36 (1):1-32.
Cheatham, C. B., and L. R. Cheatham. 1996. Redesigning Cost systems: Is Standard Costing
Obsolete? Accounting Horizons 10 (4):23-31.
Chen, C., X. Martin, S. Roychowdhury, X. Wang, and M. T. Billett. 2018. Clarity Begins at
Home: Internal Information Asymmetry and External Communication Quality. The
Accounting Review 93 (1):71-101.
Chen, C. X., S. Chen, F. Pan, and Y. Wang. 2015. Determinants and Consequences of Transfer
Pricing Autonomy: An Empirical Investigation. Journal of Management Accounting
Research 27 (2):225-259.
Chen, C. X., and T. Sandino. 2012. Can Wages Buy Honesty? The Relationship Between
Relative Wages and Employee Theft. Journal of Accounting Research 50 (4):967-1000.

108
Chen, Q. I., B. Mittendorf, and Y. U. N. Zhang. 2010. Endogenous Accounting Bias when
Decision Making and Control Interact*. Contemporary Accounting Research 27
(4):1063-1091.
Cheng, A., and R. P. Manes. 1992. The Marginal Approach to Joint Cost Allocation: A Model
for Practical Application. Journal of Management Accounting Research 4 (Fall):45-63.
Cheng, Q., Y. J. Cho, and H. Yang. 2018. Financial Reporting Changes and the Internal
Information Environment: Evidence from SFAS 142. Review of Accounting Studies 23
(1):347-383.
Chenhall, R. H., and D. Morris. 1986. The Impact of Structure, Environment, and
Interdependence on the Perceived Usefulness of Management Accounting Systems. The
Accounting Review 61 (1):16-35.
Cho, Y. J. 2015. Segment Disclosure Transparency and Internal Capital Market Efficiency:
Evidence from SFAS No. 131. Journal of Accounting Research 53 (4):669-723.
Chow, C. W., K. Haddad, and G. Totterdale. 1990. The Efficiency and Effectiveness of
Commonly Used Cost Variance Investigation Rules. British Accounting Review 22:27-
40.
Christensen, J. 1982. The Determination of Performance Standards and Participation. Journal of
Accounting Research 589-603.
Christensen, J. 2010. Accounting Errors and Errors of Accounting. The Accounting Review 85
(6):1827-1838.
Christensen, J., and J. S. Demski. 1997. Product Costing in the Presence of Endogenous Subcost
Functions. Review of Accounting Studies (2):65-87.
———. 2003. Factor Choice Distortion Under Cost Based Reimbursement. Journal of
Management Accounting Research 15:145-160.
Cinquini, L., P. Collini, A. Marelli, and A. Tenucci. 2013. Change in the Relevance of Cost
Information and Costing Systems: Evidence from Two Italian Surveys. Journal of
Management & Governance 19 (3):557-587.
Clark, J. M. 1923. Studies in the economics of overhead costs. Chicago, IL: The University of
Chicago Press.
Clinton, B. D., and A. van der Merwe. 2006. Management Accounting - Approaches,
Techniques, and Management Processes. Cost Management 20 (3):14-22.
Clinton, B. D., and S. A. Webber. 2004. RCA at Clopay. Strategic Finance 86 (4):20-26.
Cohen, S. I., and M. Loeb. 1988. Improving performance through Cost allocation. Contemporary
Accounting Research 5 (1):70-95.
———. 1990. Implicit Cost Allocation and Bidding for Contracts. Management Science 36
(9):1133-1138.
Cools, M., and R. Slagmulder. 2009. Tax-Compliant Transfer Pricing and Responsibility
Accounting. Journal of Management Accounting Research 21:151-178.
Cooper, R., and W. B. Chew. 1996. Target Costing Lets Customers, Not The Product, Set the
Price. Harvard Business Review (January-February).
Cooper, R., and R. S. Kaplan. 1987. How Cost Accounting Systematically Distorts Product
Costs. edited by W. J. a. K. Bruns, R.S. Boston.
———. 1988. How Cost Accounting Distorts Product Costs. Management Accounting 69
(10):20-27.
———. 1992. Activity-Based Systems: Measuring the Costs of Resource Usage. Accounting
Horizons 6 (3):1-12.

109
———. 1998a. Chapter 8: Strategic Activity-Based Management: Product Development. In The
Design of Cost Management Systems. New Jersey: Prentice Hall.
———. 1998b. Design of Cost Management Systems. New Jersey: Prentice Hall.
Cooper, R., R. S. Kaplan, L. S. Maisel, E. Morrissey, and O. Ronald M. 1992. From ABC to
ABM: Does Activity-Based Management Automatically Follow from an Activity-Based
Costing Project? Management Accounting 74 (5):54-57.
Cooper, R., and R. Slagmulder. 2004. Interorganizational cost management and relational
context. Accounting, Organizations and Society 29 (1):1-26.
Cornick, M., W. D. Cooper, and S. B. Wilson. 1988. How Do Companies Analyze Overhead?
Management Accounting 69 (12):41-43.
Cotton, W. D. J., S. M. Jackman, and R. A. Brown. 2003. Note on a New Zealand Replication of
the Innes et al. UK Activity-Based Costing Survey. Management Accounting Research
14 (1):67-72.
Cugini, A., A. Carù, and F. Zerbini. 2007. The Cost of Customer Satisfaction: A Framework for
Strategic Cost Management in Service Industries. European Accounting Review 16
(3):499-530.
Darrough, M. N. 1988. Variance Analysis: A Unifying Cost Function Approach. Contemporary
Accounting Research 5 (1):199-221.
Datar, S. M., and M. Gupta. 1994. Aggregation, specification, and measurement errors in product
costing. The Accounting Review 69 (4):567-591.
Datar, S. M., S. Kekre, T. Mukhopadhyay, and K. Srinivasan. 1993. Simultaneaous Estimation of
Cost Drivers. The Accounting Review 68 (3):602-614.
Davila, A., and G. Foster. 2005. Management Accounting Systems Adoption: Evidence and
Performance Implications from Early Stage / Startup Companies. The Accounting Review
80 (4):1039-1068.
———. 2007. Management Control Systems in Early-Stage Startup Companies. The Accounting
Review 82 (4):907–937.
Davila, A., and M. Wouters. 2004. Designing Cosat-Competitive Technology Products Through
Cost Management. Accounting Horizons 18 (1):13-26.
Davila, T., and M. Wouters. 2005. Managing budget emphasis through the explicit design of
conditional budgetary slack. Accounting, Organizations and Society 30 (7-8):587-608.
de Boer, L., E. Labro, and P. Morlacchi. 2001. a Review of Methods Sppporting Suplier
Selection. European Journal Of Purchasing & Supply Management 7 (1):75-89.
Degraeve, Z., E. Labro, and F. Roodhooft. 2004. Total cost of ownership purchasing of a service:
The case of airline selection at Alcatel Bell. European Journal of Operational Research
156 (1):23-40.
———. 2005. Constructing a Total Cost of Ownership Supplier Selection Methodology Based
on Activity-Based Costing and Mathematical Programming. Accounting and Business
Research 35 (1):3-27.
DeHoratius, N., and A. Raman. 2007. Store Manager Incentive Design and Retail Performance:
An Exploratory Investigation. Manufacturing & Service Operations Management 9
(4):518-534.
———. 2008. Inventory Record Inaccuracy: An Empirical Analysis. Management Science 54
(4):627-641.
Demski, J. S. 1981. Cost Allocation Games. In Joint Cost Allocation, edited by S. Moriarty.
Norman, OK: University of Oklahoma Press.

110
Demski, J. S., and G. A. Feltham. 1976. Cost Determination: A Conceptual Approach. first ed.
Ames, Iowa: The Iowa State University Press.
Demski, J. S., and R. P. Magee. 1992. A Perspective on Accounting for Defense Contracts. The
Accounting Review 67 (4):732-740.
Dhavale, D. G. 1989. Product Costing in Flexibile Manufacturing Systems. Journal of
Management Accounting Research 1:67-88.
———. 2005. Cost Considerations in Optimal Capacity Acquisition: An Option Pricing
Approach. Journal of Management Accounting Research 17:75-93.
———. 2007. Product Costing for Decision Malking In Certain Variable-Proportion
Technologies. Journal of Managment Accounting Research 19:51-70.
Dichev, I., J. Graham, C. R. Harvey, and S. Rajgopal. 2013. Earnings Quality: Evidence from the
Field. Journal of Accounting and Economics 56 (2-3):1-33.
Dierynck, B., and E. Labro. 2018. Management Accounting Information Properties and
Operations Management. Foundations and Trends® in Technology, Information and
Operations Management 12 (1):1-114.
Dikolli, S. S., and I. Vaysman. 2006. Information technology, organizational design, and transfer
pricing. Journal of Accounting and Economics 41 (1-2):201-234.
Dopuch, N., and M. Gupta. 1994. Economic Effects of Production Changes: Accounting
Implications. Journal of Management Accounting Research 6 (1):1-23.
Dorantes, C.-A., C. Li, G. F. Peters, and V. J. Richardson. 2013. The Effect of Enterprise
Systems Implementations on the Firm Information Environment. Contemporary
Accounting Research 30 (4):1427-1461.
Drake, A. R., S. F. Haka, and S. P. Ravenscroft. 1999. Cost System and Incentive Structure
Effects on Innovation, Efficiency and Profitability in Teams. The Accounting Review 74
(3):323-345.
Drury, C., and M. Tayles. 1994. Product costing in UK manufacturing organizations. European
Accounting Review 3 (3):443-470.
———. 2005. Explicating the Design of Overhead Absorption Procedures in UK Organizations.
British Accounting Review 37:47-84.
———. 2006. Profitability analysis in UK organizations: An exploratory study. The British
Accounting Review 38 (4):405-425.
Dutta, S. 2007. Dynamic Performance Measurement. Foundations and Trends® in Accounting 2
(3):175-240.
Dutta, S., and S. Reichelstein. 2010. Decentralized capacity management and internal pricing.
Review of Accounting Studies 15 (3):442-478.
Ederhof, M. 2014. Critical Access Hospitals and Cost Shifting. JAMA Internal Medicine 174
(1):143-144.
Eldenburg, L., and S. Kallapur. 1997. Changes in Hospital Service Mix and Cost Allocations in
Response to Changes in Medicare Reimbursement Schemes. Journal of Accounting and
Economics 23:31-51.
Eldenburg, L. G., K. A. Gunny, K. W. Hee, and N. Soderstrom. 2011. Earnings Management
Using Real Activities: Evidence from Nonprofit Hospitals. The Accounting Review 86
(5):1605-1630.
Evans III, J. H., R. L. Hannan, R. Krishnan, and D. V. Moser. 2001. Honesty in Managerial
Reporting. The Accounting Review 76 (4):537-559.

111
Ewert, R., and A. Wagenhofer. 2014. Interne Unternehmensrechnung. 8th ed: Springer-Verlag
Berlin Heidelberg.
Fayard, D., L. S. Lee, R. A. Leitch, and W. J. Kettinger. 2012. Effect of internal cost
management, information systems integration, and absorptive capacity on inter-
organizational cost management in supply chains. Accounting, Organizations and Society
37 (3):168-187.
Feng, H., Q. Wu, K. Muthuraman, and V. Deshpande. 2015a. Replenishment Policies for Multi-
Product Stochastic Inventory Systems with Correlated Demand and Joint-Replenishment
Costs. Production and Operations Management 24 (4):647-664.
Feng, M., C. Li, S. E. McVay, and H. Skaife. 2015b. Does Ineffective Internal Control Over
Financial Reporting Affect a Firm's Operations? Evidence from Firms' Inventory
Management. The Accounting Review 90 (2):529-557.
Fisher, M., K. Ramdas, and K. Ulrich. 1999. Component Sharing in the Management of Product
Variety: A Study of Automotive Braking Systems. Management Science 45 (3):297-315.
Foster, G., and M. Gupta. 1990. Manufacturing Overhead Cost Driver Analysis. Journal of
Accounting and Economics 12 (1-3):309-337.
———. 1994. Marketing, Cost Management and Mangement Accounting. Journal of
Management Accounting Research 6 (Fall):44-77.
Frank, W. G. 1990. "Back to the Future:" A Retrospective View of J. Maurice Clark's Studies in
the Economics of Overhead Costs. Journal of Management Accounting Research
Fall:155-166.
Gal-Or, E. 1993. Strategic Cost Allocation. The Journal of Industrial Economics 41 (4):387-402.
Gallemore, J., and E. Labro. 2015. The Importance of the Internal Information Environment for
Tax Avoidance. Journal of Accounting and Economics 60:149-167.
Gietzmann, M. B., and M. J. P. Selby. 1994. Cost-variance-based control with noisy observation.
European Accounting Review 3 (1):1-14.
Goose, D. I. 1993. Cost Accounting's Role In Computer Integrated Manufacturing : An
Empirical Field Study. Journal of Management Accounting Research 5 (Fall):159-172.
Gopalakrishnan, M., T. Libby, J. A. Samuels, and D. Swenson. 2015. The effect of cost goal
specificity and new product development process on cost reduction performance.
Accounting, Organizations and Society 42:1-11.
Göx, R. F. 2000. Strategic transfer pricing, absorption costing, and observability. Management
Accounting Research 11 (3):327-348.
———. 2002. Capacity Planning and Pricing under Uncertainty. Journal of Management
Accounting Research 14 (1):59-78.
———. 2010. Discussion of “decentralized capacity management and internal pricing”. Review
of Accounting Studies 15 (3):479-502.
Göx, R. F., and U. Schiller. 2007. An Economic Perspective On Transfer Pricing Handbook of
Management Accounting Research 2:673-695.
Goyal, M., and S. Netessine. 2007. Strategic Technology Choice and Capacity Investment Under
Demand Uncertainty. Management Science 53 (2):192-207.
———. 2011. Volume Flexibility, Product Flexibility, or Both: The Role of Demand Correlation
and Product Substitution. Manufacturing & Service Operations Management 13 (2):180-
193.

112
Granlund, M., and K. Lukka. 1998. Towards increasing business orientation: Finnish
management accountants in a changing cultural context. Management Accounting
Research 9 (2):185-211.
Gribbin, D. W., and A. H.-L. Lau. 1991. Some Empirical Evidence on The Non-Normality of
Cost Variances. Journal of Management Accounting Research 3 (Fall):88-97.
Griffin, P., C. B. Kinsky, O. Wehnert, I. Suto, and D. Canale. 2017. Operationalizing Global
Transfer Pricing: Key Steps for Translating Strategy into Practice. In EY Transfer Pricing
Survey Series.
Gupta, M., M. Pevzner, and C. Seethamraju. 2010. The Implications of Absorption Cost
Accounting and Production Decisions for Future Firm Performance and Valuation*.
Contemporary Accounting Research 27 (3):889-922.
Hansen, S. C. 1998. Cost Analysis, Cost Reduction and Competition. Journal of Management
Accounting Research 10:181-203.
Hansen, S. C. 2011. A Theoretical Analysis of the Impact of Adopting Rolling Budgets,
Activity-Based Budgeting and Beyond Budgeting. European Accounting Review 20
(2):289-319.
Hansen, S. C., and R. P. Magee. 1993. Capacity Cost and Capacity Allocation. Contemporary
Accounting Research 9 (2):635-660.
Hasija, S., E. Pinker, and R. A. Shumsky. 2010. OM Practice—Work Expands to Fill the Time
Available: Capacity Estimation and Staffing Under Parkinson's Law. Manufacturing &
Service Operations Management 12 (1):1-18.
Hemmer, T. 1996. Allocations of Sunk Capacity Costs and Joint Costs in a Linear Principal-
Agent Model. The Accounting Review 71 (3):419-432.
Hemmer, T., and E. Labro. 2008. On the Optimal Relation between the Properties of Managerial
and Financial Reporting Systems. Journal of Accounting Research 46 (5):1209-1240.
———. 2017. Management Accounting and Operations Management. edited by M. K. Starr and
S. K. Gupta. New York and London: Routledge Companions.
———. 2019, forthcoming. Management by the Numbers: A Formal Approach to Deriving
Informational and Distributional Properties of "Unmanaged' Earnings. Journal of
Accounting Research.
Herzig, N. 2012. Comment on Hiemann and Reichelstein: “Transfer Pricing in Multinational
Corporations: An Integrated Management- and Tax Perspective”.19-23.
Hiemann, M., and S. Reichelstein. 2012. Transfer Pricing in Multinational Corporations: An
Integrated Management- and Tax Perspective.3-18.
Hilton, R. W., R. J. Swieringa, and M. J. Turner. 1988. Product Pricing, Accounting Costs and
Use of Product-Costing Systems. The Accounting Review 63 (2):195-218.
Hirshleifer, J. 1956. On the Economics of Transfer Pricing. Journal of Business 29 (3):172-184.
Holmström, B. 1979. Moral Hazard and Observability. Bell Journal of Economics 10 (1):74-91.
———. 2017. Pay For Performance and Beyond. American Economic Review 107 (7):1753-
1777.
Holmström, B., and J. Tirole. 1991. Transfer Pricing and Organizational Form. Journal of Law,
Economics & Organization 7 (2):201-228.
Homburg, C. 2001. A Note on Optimal Cost Driver Selection in ABC. Management Accounting
Research 1 (12):197-205.
Homburg, C., J. Nasev, and P. Plank. 2017. The impact of cost allocation errors on price and
product-mix decisions. Review of Quantitative Finance and Accounting.

113
Hoozée, S., and W. Bruggeman. 2010. Identifying operational improvements during the design
process of a time-driven ABC system: The role of collective worker participation and
leadership style. Management Accounting Research 21 (3):185-198.
Hoozée, S. M. 2012. Chapter 9: Designing Time-Driven Activity-Based Costing Systems: A
Review and Future Research Directions. edited by F. Mitchell, H. Norreklit and M.
Jakobsen: Routledge, Taylor & Francis Group.
Hoozée, S. M., and S. C. Hansen. 2018. A Comparison of Activity-based Costing and Time-
driven Activity-based Costing. Journal of Management Accounting Research 30 (1):143-
167.
Hoozée, S. M., L. Vermeire, and W. Bruggeman. 2012. The impact of refinement on the
accuracy of time-driven ABC. Accounting Horizons 48 (4):439-472.
Horngren, C. T. 1989. Cost and Management Accounting: Yesterday and Today. Journal of
Management Accounting Research Fall:21-32.
Horngren, C. T., S. M. Datar, and M. Rajan. 2014. Cost Accounting: a Managerial Emphasis. 15
ed: Pearson Higher Education.
Horngren, C. T., S. M. Datar, and M. V. Rajan. 2015. Cost Accounting: A Managerial Emphasis.
15th ed. New Jersey: Pearson.
Horst, B. 2013a. Do Simple Costing Systems Really Exist? In Costing for Profitability, edited by
W. V. Company.
———. 2013b. When Do I Need to Review my Costing System? In Costing for Profitability,
edited by W. V. Company.
Hsu, S. H. 2011. Cost Information and Pricing: Empirical Evidence*. Contemporary Accounting
Research 28 (2):554-579.
Hsu, S. H., and S. Q. Qu. 2012. Strategic Cost Management and Institutional Changes in
Hospitals. European Accounting Review:1-33.
Hughes, J. S. 1975. Optimal Timing of Cost Information. Journal of Accounting Research 13
(2):344-349.
Hughes, J. S., and J. L. Kao. 1998. Cross-Subsidization, Cost Allocation, and Tacit Coordination.
Review of Accounting Studies 2:265-293.
Hughes, S. B., and K. A. Paulson Gjerde. 2003. Do Different Cost Systems Make a Difference.
Management Accounting Quarterly 5 (1):22-30.
Hwang, Y., J. H. Evans III, and V. G. Hegde. 1993. Product cost bias and selection of an
allocation base. Journal of Management Accounting Research 5:213-242.
Hwang, Y., and A. J. Kirby. 1994. Distorted Medicare Reimbursements: The Effect of Cost
Accountign Choices. Journal of Management Accounting Research 6:128-140.
Innes, J., F. Mitchell, and D. Sinclair. 2000. Activity-Based Costing in the U.K.’s Largest
Companies: a Comparison of 1994 and 1999 Survey Results. Management Accounting
Research 11 (3):349-362.
Ittner, C. D., W. N. Lanen, and D. F. Larcker. 2002. The Association Between Activity-Based
Costing and Manufacturing Performance. Journal of Accounting Research 40 (3):711-
726.
Ittner, C. D., D. F. Larcker, and T. Randall. 1997. The Activity-Based Cost Hierarchy,
Production Policies and Firm Profitability. Journal of Management Accounting Research
9:143-162.
Ittner, C. D., and J. Michels. 2017. Risk-Based Forecasting and Planning and Management
Earnings Forecasts. Review of Accounting Studies 22 (3):1005-1047.

114
Jacobs, F., and R. Marshall. 1999. Accuracy of Service Cost Allocations. The Journal of Cost
Analysis & Management 1 (1):45-57.
Jacobs, F. H. 1978. An Evaluation of the Effectiveness of Some Cost Variance Investigation
Methods. Journal of Accounting Research 16 (1):190-203.
Jacobs, f. H., and M. M. Ronald. 1987. A Reciprocal Service Cost Approximation. The
Accounting Review 62 (1):67-78.
Jakobsen, M. 2012. Inter-Organizational Cost Management. In The Routledge Companion to
Cost Management edited by F. Mitchell, H. Norreklit and M. Jakobsen: Taylor & Francis,
396.
Jiambalvo, J., E. Noreen, and T. Shevlin. 1997. Incremental Information Content of the Change
in the Percent of Production Added to Inventory. Contemporary Accounting Research 14
(1):69-97.
Johnsen, T. E. 2009. Supplier involvement in new product development and innovation: Taking
stock and looking to the future. Journal of Purchasing and Supply Management 15
(3):187-197.
Johnson, H. T., and R. S. Kaplan. 1987. Relevance Lost: The Rise and Fall of Management
Accounting. Boston: Harvard Business School Press.
Jones, C. L., and A. A. Roberts. 2006. Management of Financial Information in Charitable
Organizations; The Case of Joint-Cost Allocations. The Accounting Review 81 (1):159-
178.
Kajuter, P. 2012. Target Costing. In The Routledge Companion to Cost Management, edited by
F. Mitchell, H. Norreklit and M. Jakobsen: Taylor & Francis, 396.
Kanodia, C. B., Robert, and J. Dickhaut. 1989. Escalation errors and the Sunk Cost Effect: An
Explanation Based on Reputation and Information Asymmetries. Journal of Accounting
Research 27 (1):59-77.
Kaplan, R. S. 1975. The Significance and Investigation of Cost Variances: Survey and
Extensions. Journal of Accounting Research 13 (2):311-337.
———. 1988. One Cost System isn't Enough. Harvard Business Review Jan-Feb:61-66.
———. 1994. Flexible Budgeting in an activity- Based Costing Framework. Accounting
Horizons 8 (2):104-109.
Kaplan, R. S., and R. S. Anderson. 2004. Time-driven activity-based costing. Harvard Business
Review 82 (11):131–138.
Kaplan, R. S., and S. R. Anderson. 2007. Time-Driven Activity-Based Costing: A Simpler and
More Powerful Path to Higher Profits. Boston: Harvard Business School Press.
Kaplan, R. S., and U. P. Welam. 1974. Overhead Allocation with Imperfect Markets and
Nonlinear Technology. Accounting Review 49 (3):477-484.
Kaplan, R. S., M. L. Witkowski, and J. A. Hohman. 2013. Boston Children's Hospital:
Measuring Patient Costs. In Harvard Business School Cases. Harvard Business School
Publishing.
Kasanen, E., K. Lukka, and A. Siitonen. 1993. The Constructive Approach in Management
Accounting Research. Journal of Management Accounting Research Fall:243-264.
Kerremans, M., H. Theunisse, and G. Van Overloop. 1991. Impact of Automation on Cost
Accounting. Accounting and Business Research 21 (82):147-155.
Kilger, W., J. R. Pampel, and K. Vikas. 2002. Flexible Plankostenrechnung und
Deckungsbeitragsrechnung. 13 ed. Wiesbaden, Germany: Springer Gabler.

115
———. 2004. 0 Introduction: Marginal Costing as a Management Accounting Tool.
Management Accounting Quarterly 5 (2).
King, R. R., and M. Gupta. 1997. An Experimental Investigation of The Effect of Cost
Information and feedback on Product Cost Decisions. Contemporary Accounting
Research 14 (1):99-127.
Kloock, J., and U. Schiller. 1997. Marginal Costing: Cost Budgeting and Cost Variance
Analysis. Management Accounting Research 8:299-323.
Krishnan, R., J. L. Luft, and M. D. Shields. 2002. Competition and Cost Accounting: Adapting
to Changing Markets. Contemporary Accounting Research 19 (2):271-301.
Krishnan, R., and M. H. Yetman. 2011. Institutional Drivers of Reporting Decisions in Nonprofit
Hospitals. Journal of Accounting Research 49 (4):1001-1039.
Krishnan, R., M. H. Yetman, and R. J. Yetman. 2006. Expense Misreporting in Nonprofit
Organizations. The Accounting Review 81 (2):399-420.
Krumwiede, K. R. 1998. The Implementation Stages of Activity-Based Costing and the Impact
of Contextual and Organizational Factors. Journal of Management Accounting Research
10:239-277.
———. 2005. Rewards and Realities of German Cost Accounting. Strategic Finance (April):25-
34.
Kulp, S. C. 2002. The Effect of Information Precision and Information Reliability on
Manufacturer- Retailer Relationships. The Accounting Review 77 (3):653-677.
Kumar, V. 2018. A Theory of Customer Valuation: Concepts, Metrics, Strategy, and
Implementation. Journal of Marketing 82 (1):1-19.
Labro, E. 2004. The Cost Effects of Component Commonality: A Literature Review Through a
Management-Accounting Lens. Manufacturing & Service Operations Management 6
(4):358-367.
———. 2006a. Analytics of Costing System Design. In Contemporary Issues in Management
Accounting, edited by A. Bhimani. Oxford: Oxford University Press.
———. 2006b. Is a Focus on Collaborative Product Development Warranted from a Cost
Commitment Perspective? . Supply Chain Management: an International Journal 11
(6):503-509.
———. 2015. Hobby Horses Ridden. Journal of Managment Accounting Research 27 (1):133–
138.
Labro, E., and L. Stice-Lawrence. 2019. Updating Accounting Systems: Longitudinal Evidence
from the Health Care Sector. In Working Paper.
Labro, E., and T.-S. Tuomela. 2003. On bringing more action into management accounting
research: process considerations based on two constructive case studies. European
Accounting Review 12 (3):409-442.
Labro, E., and M. Vanhoucke. 2007. A Simulation Analysis of Interactions among Errors in
Costing Systems. The Accounting Review 82 (4):939-962.
———. 2008. Diversity in Resource Consumption Patterns and Robustness of Costing Systems
to Errors. Management Science 54 (10):1715-1730.
Lambert, R. A. 1985. Variance Investigation in Agency Settings. Journal of Accounting
Research 23 (2):633-647.
Lambert, R. A., and D. F. Larcker. 1989. Estimating The Marginal Cost of Operating a Service
Departments When Reciprocal Services Exist. The Accounting Review 64 (3):450-467.

116
Lanen, W. N., S. W. Anderson, and M. W. Maher. 2008. Fundamentals of Cost Accounting. New
York, US: McGraw-Hill/Irwin.
Leitch, R. A. 2001. Effect of stochasticity, capacity and lead time cost drivers on WIP and
throughput in a pull production environment. Management Accounting Research 12
(2):167-196.
Leitch, R. A., P. R. Philipoom , and T. D. Fry. 2005. Opportunity Costing Decision Heuristics
for Product Acceptance Decisions. Journal of Management Accounting Research 17:95-
117.
Lere, J. C. 1986. Product Pricing Based on Accounting Costs The Accounting Review 61 (2):318-
324.
Livnat, J., J. Ronen, and M. Swirski. 1980. Dissaggregation of Deviations in Cost Analysis. The
International Journal of Logistics Management 9 (1):71-76.
Luft, J., and M. D. Shields. 2003. Mapping Management Accounting: Graphics and Guidelines
for Theory-Consistent Empirical Research. Accounting, Organizations and Society
28:169-249.
———. 2009. Psychology Models of Management Accounting. Foundations and Trends in
Accounting 4 (3–4):199-345.
MacArthur, J. B. 1998. Cost Driver Analysis in Hospitals: A Simultaneous Equations Approach.
Journal of Managment Accounting Research 10:279-311.
Magee, R. P. 1988. Variable Cost Allocation in a Principal/Agent Setting. The Accounting
Review 63 (1):42-54.
Magee, R. P., and J. Dickhaut. 1978. Effects of Compensation Plans on Heuristics in Cost
Variance Investigations. Journal of Accounting Research 16 (2):294-314.
Maher, M. W., and M. L. Marais. 1998. A Field Study on the Limitations of Activity-Based
Costing when Resources are Provided on a Joint and Indivisible Basis. Journal of
Accounting Research 36 (1):129-142.
Maiga, A. S., and F. A. Jacobs. 2008. Extent of ABC Use and Its Consequences. Contemporary
Accounting Research 25 (2):533-566.
Mak, Y. T., and M. L. Roush. 1994. Flexible Budgeting and Variance Analysis in an Aticity-
Based Costing Environment. Accounting Horizons 8 (2):93-103.
Marcinko, D., and e. Petri. 1984. Use of the Production Function in Calculation of Standard Cost
Variances- An Extension. The Accounting Review 59 (3):488-495.
Masschelein, S., E. Cardinaels, and A. Van den Abbeele. 2012. ABC Information, Fairness
Perceptions, and Interfirm Negotiations. The Accounting Review 87 (3):951-973.
Mastilak, M. C. 2011. Cost Pool Classification and Judgment Performance. The Accounting
Review 86 (5):1709-1729.
Matsui, K. 2013. Entry deterrence through credible commitment to transfer pricing at direct cost.
Management Accounting Research 24 (3):261-275.
McGowan, A. S., and V. P. Vendrzyk. 2002. The Relation Between Cost Shifting and Segment
Profitability in the Defense-Contracting Industry. The Accounting Review 77 (4):949-969.
McNair, C. J., L. Polutnik, and R. Silvi. 2001. Cost Management and Value Creation: The
Missing Link. The European Accounting Review 10 (1):33-50.
Merchant, K. A., and M. D. Shields. 1993. When and Why to Measure Costs Less Accurately to
Improve Decision Making. Accounting Horizons 7 (2):76-81.
Mihm, J. 2010. Incentives in New Product Development Projects and the Role of Target Costing.
Management Science 56 (8):1324-1344.

117
Miller, J. G., and T. E. Vollmann. 1985. The Hidden Factory. Harvard Business Review
September:142-150.
Mintz, O., and I. S. Currim. 2013. What Drives Managerial Use of Marketing and Financial
Metrics and Does Metric Use Affect Performance of Marketing-Mix Activities? Journal
of Marketing 77 (March):17-40.
Mishra, B. K., and I. Vaysman. 2001. Cost-Systyem Choice and Incentives-Traditional vs.
Activity-Based Costing. Journal of Accounting Research 39 (3):619-641.
Modell, S. 2006. Institutional and negotiated order perspectives on cost allocations: The case of
the Swedish university sector. European Accounting Review 15 (2):219-251.
Monden, Y., and K. Hamada. 1991. Target Costing and Kaizen Costing in Japanese Automobile
Companies. Journal of Management Accounting Research 3 (Fall):17-34.
Moores, K., and S. Yuen. 2001. Management Accounting Systems and Organizational
Configuration: a life-cycle perspective. Accounting, Organizations and Society 26:351-
389.
Morratz, H., and R. Lueg. 2017. Understanding the Error-Structure of Time-Driven Activity-
Based Costing: A Pilot Implementation at a European Manufacturing Company.
European Journal of Management 17 (1):49-56.
Narayanan, V. G., and M. Smith. 2000. Impact of Competition and Taxes on Responsibility
Center Organization and Transfer Prices. Contemporary Accounting Research 17
(3):497-529.
———. 2013. Competition and Cost Accounting. Foundations and Trends® in Accounting 7
(3):131-195.
Navissi, F., and V. G. Sridharan. 2017. Determinants of Target Costing Adoption: A Research
Note. Journal of Management Accounting Research 29 (1):67-77.
Noreen, E. 1991. Conditions under which Activity-Based Cost Systems Provide Accurate Costs.
Journal of Management Accounting Research 91 (3):159-168.
Noreen, E., and N. Soderstrom. 1994. Are Overhead Costs Strictly Proportional to Activity?
Journal of Accounting and Economics 17 (1-2):225-278.
O'Brien, J., and K. Sivaramakrishnan. 1996. Coordinating Order Processing and Production
Scheduling in Order Initiated Production Environments. Journal of Management
Accounting Research 8:152-170.
Patell, J. M. 1987. Cost Accounting, Process Control, and Product Design: A Case Study of the
Hewlett- Packard Personal Office Computer Division. The Accounting Review 57
(4):808-836.
Pereira, S. C. A., and F. Mitchell. 2012. Specifying Conditions for Cost Systems Generating
Relevant Decision-Making Costs. In The Routledge Companion to Cost Management,
edited by F. Mitchell, H. Norreklit and M. Jakobsen: Taylor & Francis, 396 pp.
Pfaff, D. 1994. On the allocation of overhead costs. European Accounting Review 3 (1):49-70.
Pfeiffer, T., U. Schiller, and J. Wagner. 2011. Cost-based transfer pricing. Review of Accounting
Studies 16 (2):219-246.
Phillips, J. D. 2003. Corporate Tax-Planning Effectiveness: The Role of Compensation-Based
Incentives. The Accounting Review 78 (3):847-874.
Rajan, M. V. 1992. Cost Allocation in Multiagent Settings. The Accounting Review 67 (3):527-
545.
Rajan, M. V., and S. Reichelstein. 2009. Depreciation Rules and the Relation Between Marginal
and Historical Cost. Journal of Accounting Research 47 (3):823-865.

118
Ray, K., and M. Goldmanis. 2012. Efficient Cost Allocation. Management Science 58 (7):1341-
1356.
Ray, K., and J. Gramlich. 2016. Reconciling Full-Cost and Marginal-Cost Pricing. Journal of
Management Accounting Research 28 (1):27-37.
Reichelstein, S., and M. Hiemann. 2012. The Dual Role of Transfer prices in Multinational
Firms: Divisional Performance Measurement and Tax Optimization. The European
Financial Review (oct.).
Robinson, J. R., S. A. Sikes, and C. D. Weaver. 2010. Performance Measurement of Corporate
Tax Departments. The Accounting Review 85 (3):1035-1064.
Rossing, C. P., and C. Rohde. 2010. Overhead Cost Allocation Changes in a Transfer Pricing
Tax Compliant Multinational Enterprise. Management Accounting Research 21:199-216.
Roth, A. E., and R. E. Verrecchia. 1979. The Shapley Value As Applied to Cost Allocation: A
Reinterpretation. Journal of Accounting Research 17 (1):295-303.
Roychowdhury, S. 2006. Earnings management through real activities manipulation. Journal of
Accounting and Economics 42 (3):335-370.
Sahay, S. A. 2003. Transgfer Pricing Based on Actual Cost. Journal of Management Accounting
Research 15:177-192.
Sansing, R. C. 1999. Relationship-Specific Investments and the Transfer Pricing Paradox.
Review of Accounting Studies 4:119-134.
———. 2014. International Transfer Princing. Foundations and Trends in Accounting 9 (1):1-
57.
Schoute, M. 2011. The Relationship Between Product Diversity, Usage of Advanced
Manufacturing Technologies and Activity-Based Costing Adoption. The British
Accounting Review 43 (2):120-134.
Scott, D., E. Labro, M. Bolognesi, S. Wellman, C. Maher III, and C. Penrose. 2018. The Impact
of Electronic Medical Record Implementation on Labor Cost and Productivity at an
Outpatient Orthopedic Clinic. Journal of Bone & Joint Surgery 100A (18):1549-1556.
Shackelford, D. A., and T. Shevlin. 2001. Empirical Tax Research in Accounting. Journal of
Accounting and Economics 31:321-387.
Shank, J. K. 1989. Strategic Cost Management: New Wine, or Just New Bottles? Journal of
Management Accounting Research 1 (Fall):48-64.
Shank, J. K., and V. Govindarajan. 1988. The Perils of Cost Allocation Based on Production
Volumes. Accounting Horizons 2 (4):71-79.
———. 1992. Strategic Cost Management: The Value Chain Perspective. Journal of
Management Accounting Research 4 (Fall):180-197.
Sharman, P. A., and K. Vikas. 2004. Lessons from German Cost Accounting. Strategic Finance
(December):28-35.
Shin, J., K. Sudhir, and D.-H. Yoon. 2012. When to "Fire" Customers: Customer Cost-Based
Pricing. Management Science 58 (5):932-947.
Shroff, N. 2017. Corporate Investment and Changes in GAAP. Review of Accounting Studies 22
(1):1-63.
Simons, R. 1987. Accounting Control Systems and Business Strategy : an Empirical Analysis.
Accounting, Organizations and Society 12 (4):357-374.
Sprinkle, G. B. 2003. Perspectives on Experimental Research in Managerial Accounting.
Accounting, Organizations and Society 28:287-318.

119
Subramanian, U., J. S. Raju, and Z. J. Zhang. 2014. The Strategic Value of High-Cost
Customers. Management Science 60 (2):494-507.
Suh, Y. S. 1987. Collusion and Noncontrollable Cost Allocation. Journal of Accounting
Research 25:22-46.
Trenchard, P. M., and R. Dixon. 2003. The clinical allocation of joint blood product costs.
Management Accounting Research 14 (2):165-176.
Trentmann, N., and E. Minaya. 2018. Finance Chiefs Say Too Much Data is Making it Harder to
Keep on Top of Risks. Wall Street Journal, July 19, 2018.
Tse, M. S. C., and M. Z. Gong. 2009. Recognition of Idle Resources in Time-Driven Activity-
Based Costing and Resource Consumption Accounting Models. Journal of Applied
Management Accounting Research 7 (2):41-54.
Ulrich, K. T., and S. Pearson. 1998. Assessing the Importance of Design thorugh Product
Archaeology. Management Science 44 (3):352-369.
Van den Abbeele, A., F. Roodhooft, and L. Warlop. 2009. The effect of cost information on
buyer–supplier negotiations in different power settings. Accounting, Organizations and
Society 34 (2):245-266.
van der Merwe, A., and D. E. Keys. 2002. The Case for Resource Consumption Accounting.
Strategic Finance 83 (10):31-36.
Vaysman, I. 1996. A Model of Cost-Based Transfer Pricing. Review of Accounting Studies 1:73-
108.
Vollmers, G. L. 1996. Academic Cost Accounting from 1920-1950: Alive and Well. Journal of
Managment Accounting Research 8 (183-199).
Wagner, S., and L. Dittmar. 2006. The Unexpected Benefits of Sarbanes-Oxley. Harvard
Business Review April:133-140.
Wei, D. 2004. Inter-Departmental Cost Allocation and Investement Incentives. Review of
Accounting Studies 9:97-116.
West, T. 2013. Costing Mistakes 101: Not Re-Evaluating your System. In Costing for
Profitability, edited by W. V. Company.
Williams, D. J., and K. J. O. S. 1983. A Unique Procedure for Allocating Joint Costs from a
Production Process? Journal of Accounting Research 21 (2):644-645.
Woods, M., L. Taylor, and G. C. G. Fang. 2012. Electronics: A case study of economic value
added in target costing. Management Accounting Research 23 (4):261-277.
Wouters, M., J. C. Anderson, and F. Wynstra. 2005. The adoption of total cost of ownership for
sourcing decisions––a structural equations analysis. Accounting, Organizations and
Society 30 (2):167-191.
Wouters, M., and S. Morales. 2014. The Contemporary Art of Cost Management Methods during
Product Development. 24:259-346.
Wouters, M., S. Morales, S. Grollmuss, and M. Scheer. 2016. Methods for Cost Management
during Product Development: A Review and Comparison of Different Literatures.
26:139-274.
Wynstra, F., A. v. Weele, and B. Axelsson. 1999. Purchasing Involvement in Product
Development: A Framework. European Journal Of Purchasing & Supply Management 5
(1):129-141.
Yoshikawa, T., and R. Kouhy. 2012. Continuous Improvemnt and Kaizen Costing. In The
Routledge Companion to Cost Management, edited by F. Mitchell, H. Norreklit and M.
Jakobsen: Taylor & Francis.

120
 Zimmerman, J. 1979. The Costs and Benefits of Cost Allocations. The Accounting Review 54
(3):504-521.

121

View publication stats