l-

After reading tizis chapter, you should have an appreciation

oi

the followino,

I ttte importance of measurement

E ,t " nominal, ordinal, interval and ratio scales of measurement
! ttre permissible operations of scales
I tt " difference between fundamental, derived and fiat measurements
! what is meant by reliability and accuracy in measurement
! me"sure.ent in accounting
I measure.ent issues for auditors.

+
+'E.

rrl
Measurement is a significant pan of scimtific inquiry. Measurements are made" as
demonstrated in accounting because quantitative dau can impan greater information
than qualiutive data in many instances. Be@us measurcment of attibutes reported
in accounting report (e.9. asstr, incone and liabilities) is an imponant function in
accounting it is wordrwhile for us to e(amine measurement theory and to oudine a
number of basic measurment assumptions in accounting. .

1,
t,
I

.i-

IMPORTANCE OF MEASURTMENT
Campbell, one of the first to deal with the issue of measurcment, de6ned measurement
as'the assignment of numerals to represent propenies of material systems other than
numbers, in vinuc ui dr< iaws guvertrirr! iircc pruprii<u'.: Sacvttu, a luicri iir<rlili i.. ilit
to measurement as the 'assignmmt of
area of measurement in the social sciences,
nurnerals to obiects or events according to rules'J Campbell makes a distincrion benveen
systems and the propenies of those systems. The'systems'in Campbellt definition are
what Stevens slls 'objects or events'. These could include houses, tables, peoplc assets
or distance travelled. Propertis are the specific aspects or charactetbticE of th systems,
such as weigh! length, width or colour. We always measure properties and not the sycterns
rhemselves. ln this re$peq Campbellt defrnition is more precise than Stevens's.
Campbell's definition requires numerals to be assigned to propenies according to
the laws governing the propnies, whereas Stevens's definition requires only that the
assignment b done 'according to rules'. Sterling obiedts to the broadness of Stevens s
definition, arguing 'One needs restrictions upon the kind of nrle that can be used'.3
Othenvise, any assignment of numbers can be called measurement. ln the usual
understanding of measurcment, semantic rules (operationd defnitions) ue devised
and used to link the formal number system wirh the property (of obiects or evenu) to
be measured. When the semantic rules assign numbers to obiecB or events in such a
way that the relationships among the obiects or events (in respect ofa given ProPerty)
correspond to mathematical relationships, a scale has been established and the property

reffi

is

.l

said to be measured. Stevms states:

When rhis conespondence between the formal model and is empirical counterPart
is close and tight we 6nd ourselves able to discover truths about maners of fact by
examining the model iself.a

Under this view, the measurement process is similar to the aPProach to theory
formulation and testing mentioned earlier' A statemenL exPressd mathematicalln
is advanced. Semantic rules (operations) are devised to (onnect the symbols of the
statement to panicular obiects or events. When it is demonstrated that the relationships
in the mathematical staternent coftelate with the relationships of the obiecrs or events,
then measurement of the given aspect of the obiects or events has been made.
In accounting we measure profit by first assigning a value to capital and then

calculating profit as the change

in capital over the period

after accounting

for all

eonomic events that affecr the wealth of the 6rm.

e @-.

SCALES
Uvery measurement is made on a scale. A scale is created when a semandc rule is used

to relate the mathematical statement to obiects or events'

The scale shows what information the numbers represent, thus gMng meaning to
the numbers.'l'he type ofscale qeated depends on th semantic rules used. Accordlng
131

PART

Theory and accounling ptJritice

to Stevens, scales can be described in general terms as nominal, ordinal, intewal or

ratio.s 'Ihese dassifications were arrived at by e(amining the mathematical goup
suucture ofscales. The mathernatical structure is d*ermined by considering the ki-nd
of
transformation that leaves rhe structure of the scale invarianc i.e. unchangid.

Nominalscale
In the nomina.l scale, numbers are used only as labels. The numbering of football
players is an example given by Stevens.
Many theorists obiect o the norninal scale as representing measurment. Torgerson
states:

ln measuremnt, as we use the term, the number asigned refers to the relative
amounr
or degree ofa property possessed by the object, and n--ot to the objea
itself whereax in
the different nominal scales, the numbers refer to the obiects or
classes ofobjects: it is
the object that is named or clasi6ed.6
The nominal scale sirnply represenu classification, which is not
what measurement

is considered to be in the ordinary usage of the term. As Torgerson polnts our,

rneasurement refers to properties of obiecls, whereas in the nominal

scale the numbers
often denote the obiects themselves, ruch as numbering or naming players
in sponing
teams. The maior property the numbers have is to identifi playeis

accounting sFtem, the dosest we have

assets and liabilities into different dasses.

or oblects. In thto the nominal scaie t the classification of

Ordinalscale
An ordinal scale is creakd when an operation ranla the obiects in question
with respect
to a givenproperty. For example, suppose a cenain investorias
tluee feasible inuestment
opponunities for a givm amount ofmoney to invest. They are ranked
l. 2, 3 according to
their ner present values, wirh the highest ranl<ed as I and the lowest
as 3. The opention
(calculation of net preseni value) gives rise to an ordinal
scale. which is the set of
numbers refening to the investment altematives. The numbers
iniicate the order of the
size ofthe net present value of the options and, thereforg
their profiability,
A weakness of the ordinal sole is that the intervals between the
numbers (l to 2,
2 to 3 and I ro 3) do not tell us anything about the differences
in the quantity of rhe
propery they rprErenl In our erample, in terms ofthe
asp.a *"arur"d (net present
be
very
dose
to
oprion
l,
and
-ay
option a rray be consideraLly less
than option 2. Another wealaness is that the numbers do not signiry
how much, of ttre
anribute the obiects posess.
Torgerson argues that some ordinal scales have a ,natunl
origin,, that is, a natural
zero poinrT Applied to our example of nnhng investrnent
altema-tives, the narural zero
poht could be a neutal point where in one direction are all
the o(I)eced profitable
dtemativ* and in the other direction are the expected unprofiabi"
irr"s, rr,e nuo'Uen
aesigned to the options on one side of the zero poinr wouia
n"* po.itir" ,igns and, on

:."!"ry,i": i

the other. negatira signs.

Interval scale
The interval -scale imparu morc information than the ordinal
scale. Not only is the
nnking of the obiects known with respect to the given property, but the
disrance
between the intnra.ls on rhe scale is eiual and f.n-or. a
*t..ii
point also
exists on the scale. An example is the Celsius scale of temperature. ""ro
Equal intervals of
CHAPTER

Measu.ement theory

tempenrture ,ue noted by equal volumes of opansion with an arbitrary zero point
agreed on for the scale. 'Ihe temperature differential is divided baween fteezing and
boiling into 100 degrees, with the fuezing point arbitrarily sa at zero degrees. If the
temperature of two different rooms is measured with a Celsius thermometer and gives
readings of 22 degrees and 30 degrees, we can say not only that rhe second room is
hotter, but also that it is 8 degrees higher in temperature. The differences between the'
numbers can be transiateci direaly to represent the differences in the characteristic of
the obiecrs.
The weaknes of the interval scale is rhat the zero point is arbitrarily etablished.
For example, suppose we wer to measure the height ofa group of men on an interval
scale and assign a number to eadt according to his height with respect to the average
of thc qrnrrn Thq avqr4.ge rEryes4nts t-he zero poinr cn .!'.e scgle. !f .{ !s 3 centirn+Jeg
above the arrerage, then we would asign him the number +3; and ifB is 5 centimetres
!g!ow.t!g avg|age, tben rrye qro-uld a$!ig4 fuim t]r.g num]er ,.5. On this scale, we do not
know how tallA or I is in an absolute (actual) sense. B may be the shortest man in the
group, but the group may consist of tall basketball playen.

Mattesich irentions standard cost accounting

as

one xample where the intennl scale

is used in accounting.s Th standard may be based on theoretical, average, pradical or

normal performance. Because the choice is more or less arbitrary, the calculation of
standards and nariances gdnerates an intewal scale. If rhe variance is zero, this sigrifies
neuuality, but this point is arbiuarily seleaed.

Ratio scale
A ratio scale is one where:
the rank order of the obleas or events with resPect to a given property is known
the inter ls bween the obiects are equal and are known
a unique origin, a natural zero poinl exists where the distance from it for at least one
obiecr is known.
The ratio scale conveys the most information.
The measurement of length is a good orample of a ratio scale. When A is l0 meEes
long and B is 20 metrcs, we can say not only that B is 10 metres longer but also that it
is twice as long as A. The ratios of the numben are also direcrly interpretable as ratios
of the quantiries of the propeny measured. Thus, it makes sense lo say that A is half as
long as B or is nrice as long as A, whereas we cannot sily that 40 degrees C-clsius is
twice as hot as 20 degrees Celsius.
An example of the ratio scale in accounting is the use of dollars to represent cost and
value. lf asset A cost $lo oo0 and asst I aost t20 000, we can state that I cost ttYice as
much as A. A natural zero point exists, because 0 denotes absence of cost or value, iust
as 0 for length means no length at all.

.
.
.

-* @a

PERMISSIBLE OPERATIONS OF SCALES

One reason for discussing scales is that ceftain mathematical applications are permissible
only for differmt types of scales. The ratio scale allows for all the fundamental
arithmetical operations of addition, subtraction, multiplication and division, and
also algebra, analytic geomary calculus and statistical methods. A ratio scale remains
invariant (fi:<ed) over all transformations when multiplied by a consumle For qomple,
consider the following:

X'=cX
PARr

l.a-,

--,lli

2 Theorv and accounting Ptactice

IfX represents all the points on a given scale, and each point is multiplied by a constant c, the resulting scale
will also be a ratio scale. The reason is that the structure of
the scale is left innariant that is:

o the rank order ofthe points is unchanged

. the ratios of the points are unchanged

. rhe zero point is undranged.

'Ihis means that if we measured the rength of a roorn and found it

to be 400 centimares
and then converted rhe 4OO centimetres to 4 meues by multiplying by the constant
l/100,

we can be assured that the length ofthe room is unchanged even though the
number
representing length has changed. This is the same point we make in chapter O regarding
the convenion ofhistorical cosr of, say, $ 100 000 ofequipment under rhe nominal dollai
scale to the purdrasing power of the dollar scale by applying a conshnr, say, t2o/f00,
to
derive

$ 120 000. The $ r20 000 is stilt historical cost.

Th9 iryalance of a scale permits us to know re enent to which a rheory or rule
remairu basically the same, even though the scale is ergressed in different units, such
as from centimetres to metres or from nominal dollars to constant
dollars. An invariant
transformadon ofa ntio scale will leave inkct the same general form ofthe relationship

of the nariables.
. Without riance, it is possible to find that X is twice as long as y when measured
in centimeues, but three times as long when mealured in rnetr;-s. In accounting the
scale for current cost
riant ftom that of historical cos! because the attribuies to

in

is

be measured are different. When machine r{ is measured under historical

cost it may
be $90000, but when measured under cunent cost it may be tt10000..Ihe
unit of
measure, the dollar, is used in both cases but the scales are different
they are variant.
But dnnging from the nominal dollar scale to the purchasing power
of the dollar scale

for rhe same attribute (historial cost or current cost) reaves tie structure invariant.
with an interval scale, not dr arirhmeticar opentions are permissibre. Mdition and
subtraction can be used wirh rspect to the panicular numbers on the scale as
well as
the intenrals. Howvel multiplication and division cannot be used with
reference to
the panicular numbe^, only to the inteffals. lo The reasons are due to the
conditions of
invariance. An intervar scare is invariant under any linear transformation
of the form:

K=cX+b
The uansformation of one interval scale for measuring a specific_ property
to another
intenral.scale for measuring the same property is made by multiplying
eachpoint of the
fist sole X by a consant c and adding to it a constant b. The re"son for a is that
erc
is no absolute zero point on an inten"al scale. For examplq to
transform

from Celsius

o lahrenheit
s.p=tut"
The 9/5 is used because

temperature, we multiply each deglee'.by 915 and add

12.
the celsius scare has l0o d-{ees as opiosid ro 180 degrees ior
Fahrenheit and 32 is added because that is the freezlng poini ior the
latter scale.
The conditions ofinvariance show that we can murtifry and
divide with respect to the
intewals, but thes arithmeticar operations cannot be-used for the paniculai
numbers
on the scale. To illustrate, corrider the following rransformarion:

X=X+ l0
Consider the obiects on points 3 and 6 on scale X. Transforming to scale
)(, we now
have 13 and 16. The ratio of 13 to 16 is not rhe same as the raiio of 3
to 6 because
of the addition of the constanr. Murtiplication and division (i.e. ratios) are therefore
not permissible for rhe paniorlar numbers. Tfiuq if Robyn receives
90 poins on her
accounting xam and Maria receives 45 points, we cannot say that
Robyn knows twice
as mudr as Maria regarding the subiea maner of the exam. ih"
is that there is

r.*n

CHAPTER

Measurement theory

1?7

no natural zero point for the exam for 'no knowledgd- Errcn if a student receirres '0' on
the eiram, we cannot say that he or she has no lsrowledge ofthe subiect maner. In this
examplg what we can say is that Robyn passed the oom and Maria failed the oram, but
we cannot infer comparativdy the amount of knowledge to the numbers. Likewise if
the quantity variance is $5000 favourable as opposed to the previous month's variance
of $ l0 000 favourable, we cannot say that the use of materials this month is only half
as mcient as in &e pieceding month.
With ordinal scales, none ofthe arithmetical operations can be usd. We cannot ad4
subtracg multiply or divide the numbers or the intervals on the scale. Ordinal scales,
therefore, convey limited information.

*:"

@*

TYPES OF ME,A.SUREMENT
As we said eadier, rhe mea3urement proces is similar to the scientific approach of
theory cbnstructictn iiid tStihJ. Oiir disciriiioii of riles relates to thC question of the
constuction and implemmution of theory. There must be a rule to assign numbers
before there cin be measurement. This rule is usually a sa of operations whidr must be
devised for the given task. The formulation of the nrle gives rise to a scale. Meagurement
can be made only on a scale.
The question of testing of a theory relates to the question of the different kinds of

measuremmt. Campbell mentioned two kinds: fundamenal and derived.ll Recall

tbat in Campbell's definition of rneasurement he stated that the numben are assignd
according to the 'laws' gowming the property. For Campbell, masuremen$ can
take place only when there are confirmed empirical theories (laws) to support the
measlrements. A further type of measurement, fiat measuremenl was mentioned by
Torgerson as being additional to the fundamental and derived measuements discussed
by Campbell. All three measurements are discussed on the following pages.

Fundamental measurements
A fundamental measurement is one where the numbers can be assigned to th proPerty
by reftrmce to natural laws and which does not depend on the measurement of any
other nariable. Propenies such as length, electrical resistance, number and rolume are
fun&mentally measurable. A ratio scale can be formulated for each of these properties
on the basis of laws relating the differmt measures (quantities) of the given propeny.

'Ihe interpraation of the numbers depmds on the confirrned empirical theory that
govems the measurement oPntion.

it tunrs out, the fundamental propenies are additive. Because of this, it is simple
to find physical parallels to dre oPerations of arithmetic.r2 For e:rample, adding the
length of obiect X o rhe length of obiect Y is panlleled by the actud operation of
placing two straight rods md to end, with one rod having the same length as X and
As

the other the came length as Y. We can physically determine the total length of X and
L Because of these phFicat parallels, the scientist can simply perform the medtanical
opentions of rnathematics without having to conduct atr ogeriment for length.

Derived measurements
According to Campbell, a derived measurement is one that depends on the measutment
of two or more other quantities. The measurement of density is an orample. lt depends

thl

of both

mass and volume. Derived measurment operations

depend on known relationships to fundamental propenies. They are based on a
confirmed empirical theory relating the given property to other propenies. Mathematical

on

i;':;,',:iig
:.i.:: -:':,:
,l

PART

measurement

Theo.y and accounting praclice

operations can be performed on the numbers. from derived measurements because of

the parallel mathemari@l and phFical operations ori rhe fundamental propenies.
It has been pointed out that therc are measuremenB such as tempeftrture, that depend
only on one rather than two or more other measuremenE,r3 To mearrure temper:rture,
we need only to measure pressure, volume or electrical resistance. However, even in
these cases the measurement is based on natunl laws.
Today, because natural scientists are aware of so rnany

loown relationships among

physical properties, they can simply derive measuremenB based on a few fundamental
properties. But this cannot be said of social scientists, because there is no agreement as
to what dre fundamental properties are in the social sciences- ln accounting an oample
of a derived measurement is profit. It is derived from the addition and subtraction of
income and expenses.

Fiat measurements
It is truical in the social sciences, and in accounting to use an arbitrarily esrablished
definition to relate cenain observable propenies (variables) to a given concepL without

having a confirmed theory to suppon rhis relationsbip.

For orample, in accounting we do not know how to measure the concept of profit
directly. Rather, we assume that the variables of revenues, gins, expenses and losses
are related to th concepr of profit and can therefore be used to give us an indirect
measure of profiL We use an arbitrary definition to relate the variables to the concept.
In this instance, we consider the algebraic sum of the measurement of the variables to
be a measure of profit.
However, under Campbellt stringent da{cificatioo rneasurement can be made only
if- confirmed empiriol theories exisrt to suppon fiem. Under Campbelt,s requiremeni
therl many of the measuements in the social sciences, including our measurement of
profil cannot be aonsidercd measuremmts.'
ln order to iustifi most of the measurements in the social sciences, Torgerson
argues that one other category of measurement should be added to Campbelt,s
list measurerrent by fiat.lt (fib, means decee, edicu) Such measurimenc would
encompass those based on arbiuary definitions (e.g. the measurement of profit in
accounting). However, Torgerson points out that the.ffittrilftibhfi.iditlr.nrd3'Urt{nent,
b5tr.Ei[lbecause it is not based on confirmed rheory iffiGglmrrous rurys in which the.
sg|gpg1,be c-olu1rucad. In accounting for erample, the rarious accounting Etandards
boards determine accounting scale by fiaL not by reference to confirmed measurement
theories. thereforg there are many measurement alrematirres so confidence in anv
pafiicular scale may be low Coming back to our earlier sample, do we know for
example, that the specific way wi measure profir is valid? lt may be one ofa hundred
waF to masure profit and as long as the paniorlar way we measure it is not based on
confirmed rheory there is no good reason for confdence in its results.
oneiifrhe i'sbns for a deasuimedt approactr to accorinting theory formulation is the
hgpe tbatifaccountingrhcory.on be empirically tet4 rhn iutead oifiat
-"."ure."nt
\de.can fiave fundarnmtal measuremenB. One can be
iustifiabiy more confident in
fundamenal measurtmems than in fiat measurementc as a measurement tool.
To test for the validity of their measur6, social scientists have attempted to relate
the property under study to otler variables to see if they are meaningful. For erample
if we wanted to measure people's arithmetical ability, we might choose to make ihem
sit an arithmaic t6t. Howa/er, there is no con6rmed empirical theory to
iustify our
test and we therefore make assumptions rvhen we establish the measurement scale
(i.e. when we set the number and type of items in the test and the form of the test). We
CHAPTTR

Measurement theory

.1:r9 .,

can prdict that, given a large number of pmple, those who scorc high on oul te$ will
dso perform well in a university maths cous. The correlation between scores on the
test and the grades received in the univenity course would be one way to validate the

particular mearuremenr opention (i.e. ou! test). In this way, asuming a significantly
high positive correlatiorl we can have some confidence in the validity of the given
measurement oPention.

-78tr.

RELIABILITY AND ACCURACY

What is meant by the reliability of a measurement or the accuracy of a measurement?
To answer this question, we should state 6rst that no measurement is ftee of error
o<cept counting. We can count the number of drain in a certain roorn and be exaaly
correct. But xcept for counting all measurements inyolve erlors.

Sources of error
The sources df eaor in measurement include the following which are not mutually
exclusive.

Ma^suftnent oprati,'a snuil inpruLdy. The nrle to assign numben for a given
propeny usually consists of a sa of operations. A set of operatioru rnay not be stated
prccisely and therefore may be interpreted incorrecrly by the measurer. For examplg
the calculation of pro6t involves numen)rE opentions, sudr as cost dassifications
and allocations betrveen assets and openses which are often interpreted differently by
different accountants. Another reason is that frequendy the fit of the mathematical
operation does not match well the acnral relations ofthe properties to be measured.
MoaflteL The measurer may misinterpm the rule be biased, or apply or read the
instument incorrectly. For qample, if ten people measured the length of a cmain

room, there would probably be ten different resulc, whidl may all be dose, but
still at variance with one another. One concem in accounting is that managers have
cenain biass to inceas the rccorded profit or assa base and then place pressure on
the accountant to bias the accounts.
lnstrumatL Many operations call for the use of a physicat irtslrummt such as a ruler
or thermometer or barometer, whidr may be flawed. There is potential for error even
when the instummt is not a physical tool but is, for examplq a chaO graplr, table
of numbers or a price index" For instance, some consider the CPI for general price

levtl adiustments to be defectfue.

Eavdtonment.The eating in whidr the measurcment opention is performed can afiect
the result. For erample, weathr conditions may affect the insuument or lhe measuer.
More generally, noise may distracr the meirurr or, in accounting pressure from
manatment may afiect dre accountantt decisions. If the pressure causes bias by the
accountant, the'ero/ is deliberate and non-random. tf the pressure (e.g from heavy
workload) caus6 concen&rtion lapees and distractioo the source of error can be
labelled 'mvironmental', Random errors are often caused by environmmtal factors.
Another factor is the environmert in whidr the firm's managemmt is operating in.
For oample the manager may be paid his or her bonus according to the amount of
proit eamed or the cost of debt funds mry be determined by the amount of gearing

(asets/equity).
Awfuttte utrlear. What is to be measured may not be dear, esp*ially if the
masurement involves a concpt which cannot be measured directly. For orample,
cuppose we want to measure the mechanical ability of people, which is not a
direcly obsewable property. What do we look at to measure? Or suppose we wish

Theory and accounting practice

"tto

to measure fte 'masculinity' of eadr boy in a given group. First of all, rhe atuibute
is difficult to defne. Measuremenr of it can only be indirecly infered from a
variety of responses, The problem of vagueness of the attribute is not uncommon
in accounting. What is rhe value of a non-(urnent asset? ls it the present value,
acquisition cos! curent cost or selling price? Given that a maioi purpose of
accounting is to reflecr 'values', it is important to dearly define the anribute ,value,.
Is it value in use, value in s(change, or some other aruibute that the accountant
should measure? The problem lies in defining the attribute to be measured, not the

measurement method ger se.

Rish anil uncerainty. This relates to the distribution ofre ms on a tangible asst. For
q:<ample, firture retums on tangible assets such as plant and equipmenl are rislcy but
they are (more or les) homogeneous and prices obsenrable. That is, when pricing
the asset one may under- or overestimate rhe amount of raurns but rhe distribution

of retums is more or less known (i.e. the estimate of value is risky but the risk is
couiparably narow). However, inungible asse$ hce the problem of risk as well as
uncertainty. Risk exists because the amount of reurns is unknown, and uncenainty
means that we face a reladvely unknown distribution of returns. This is usuallv
caused by highly uncenain retums from inungible investmena (e.g. human capitai
research and development and marketing), and raurns from these investments that
vary greatly across 6rms and industries.
If all measurements orcept counting inherently involve enors, then how can any
statement that includes a measurement b regarded as true? The probtem is rhat many
orpect perfection when there 6nnor be any. What we need is to establish limir o]
acceptable error. If any measurement falls within these timis then it can be considered
true and fair in accounting te.ms.

Reliable measurement
It is often required that before elemens such as assts, liabilities, income and enpenses
are recognised in fnancial shrements, the elemms shoutd be capable of reliabte
measurement (e.g. IAS 8/AASB 108 Accoun hg policies, Cl1ulng.s in kuuating Fsrtmates
arul Eros, paragraph l0(b)). What is meant by a reliable maluemmt? Iteliability
refers to the proven consistency of either an operation to produce satisfactory resuls
or the reults (the numbers) rhemselves for a panicllar use. In statistics, Rliability
demands that measurements be repeauble or reproducible, thereby demonsuating
.:.
their
The notion of reliability incorporates two aspec$: the accuricy and certaing of
measuremenL and the representative faithfulness of disdosures in relation to the
underlying economic transactions and events. The measurement aspect concems the

consistency.

precision of measurement.
The term precision' is often used in two contexts. Firsr, it may refer to a number, in
which case it is opposite o the notion ofapproximation. For oamplg the number 90.4
is more precise than 90 ifthe real score is 90.44. Secon{ it can refer rc a measurement
opention, in which case ii relates ro:
o the degree of refinement of the operation or io performance
the agrement of the results among repeated use of the measurement operation as
applied to a given property.
this last meaning is essentially lhe same as that of retiability. Ilrtting together rhe
two tenns, we can say that the reliability ofa measurement penains to rhe precision
with which a speciFc property is measured by use ofa given set of operations.

CHAPTER5 Meaiurcment theory

Accurate measurement
Althougb a measurement procedure may be highly reliable giving very precise results,
it may not produce accurate resule. A cenain rifle in the hands of an er<pm sports
shooter may be highly reliable in enabling successive shots ro be placed close together,
but if the sight is not properly aligned, those shots will not be around the bullseye. '
Consistency of resulu, precision and reliability do not necessarily lead to accuracy. 'Ihe
reason is that accuncy has to do with how close the measurement is to the 'true value'
of the anribute measure the tullseye', so to spex[ts
Fundamental properties, sudr as the length of an obleG can be daermined to be
acurate by comparing the obiea with a standard that represents tjrue value. We can, for
oamplg use a ruler as a rcpresentadon of rhe standard. For many years, a platinumJ--l
---^
ul<- -.-tllutulu uar K<pt ur fdll! t<Pt<wul<u ur< lktrl|rcru rll<ui. tlr rroJr .L
was de-fined as the length of the path mvelled by light in a vacuum during a time
interval of 1/(299 792 458) ofa second.
Ttre problem is that for many measurements the true v"lu is not known. In order to
determine accrracy in accounting we need to know what attribute we should measure
to adrierre the purpose of the measurement. The obiective of accounting mentions
the 'usefulnessl of the information. Acqrracy of masurements therefore rdates to the
pragmatic notion of usefulness, but accountants are not in agrement as to what the
specific quantitative standards are ihat aie implied. We should note, however, that
repetition of operations does not erxrurc accuracy. We can calculate the cost of the
invenory by FIFO and repeat the calculation a hundred times and ga the same anslver,
but that does not mean the answer is accurate, except in the sense of checking for
arithmetical enor. We can also be very precise in our calculation to derive a figure of
$ r 018 412.18 and still not necessarily be accurate. Instead of using the term 'acanracy',
whidr is so oftm understood to mean arithmaical precision, it may be prudnt to use
the term of the social scientists, 'validity'.
The following interview provides a perspecrive regarding the meaning of reliability
and relerrant measurement. Imponantly, note the comments on the hienrdry of
reliability in nleasurements, increased complority, the new role of auditors and
unitrersity courses.

.l
t\, |i\.\lr)\\l
lLr

+--.

The impacl of the accounting profession's movement toward fair value reporling in financial
3tatements: an interyiew with Theresa Ahlstrom, Long lsland Office Managing Partner,
KPMG tTP.

lnnviewed by Pattick A. Caafuna, The

St.

Mn's

Peter

l. Tobin

College of Suiness,

Unive6ity

KPMG LLP. the audit, tax and advisory firm, tums knowledge into value for the benefit of
iB clien6, people, communities and the caPital markeE. ls prcfessionals wo* together to
provide cliintr access to global suppo( industry insighs, and a multidisciplinary range of
iervices. KPMG LLP (www.us.kpmg.com) is the U.S. member firm of KPMC lnternational. In
1993, KPMG lnternational was the first muhidisciplinary professional services organization
to establish itself alon8 industry-specific lines of business. This enabled KPMC to tailor
services and strategies to the needs of clients across a range of global industry markets.
I(PMG International's member firms have 103,000 profesionals, including 6,700 partners,
in 144 countries.
Theresa P. Ahlstrom is KPMC's Long lsland Office Managing Partnerandthe Northeast Audit
Quality Support Panner, supporting the Northean Audit Area Risk I Management Partner,
and playing a key role in the development and orecution of the area's quality enhancement
initiitives. She started her career at KPMG, LLP, in 1982 as a preprofessional in the firm's

' '.' .fi2

PART

Theory and accounting p.actice

,"

National Department of Professional Practice (DPP), and joined the audit professional staft
of the Long tsland office in 1983. In 1993, after a two-yeai rotation in Dpp and the Office of
Ceneral Counsel, she was admitted into the partnership.
Ms. Ahlstrom has served as lead engagement partner for numerous clients in the healthcare,
biotech, consumer and industrial market, and non-profit industries. Because of her technicai
experience and client service record, she was designated an SEC Reviewing partner,
Professional Practice Partner, Employee Benefit Resource Partner, Primary Campusiecruiter,
and National Training Instructor;
Craduating summa cum- laude with a B.S. degree in accounting from St. fohn's University
in 1983, she was inducted into the YWCA Academy of Women Achievers in 1998 and the
Long lsland's Top 50 Women Hall of Fame in 2000, 2001 and 2003. Ms. Ahlstrom is also
very active in numerous community activities and professional associations, and receivecl
the 2003 Long lsland Fund for Wonren and Cirls Achievers Award. She resides in Sor-rth
Huntington, New York, with her husband Bob and their two young sons.

Q: How do you think fair value affects the reliability and relevancy of the financial

A:

statements?
There have been grumbles within Corporate America for well over a decade that financial
statements are irrelevant to iinancial analysts, in light of the current techniques analysts
use to value the health and projected financial perfornrance of an entity. The objeciive
of fair value accounting on the other hand, provides users of firrancial statemens with
a clearer picture of the current economic state of a company, making a conrpany's
financial statements more useful or "relevant" in the marketplace. Clearly, historicil cost
accounting, while perhaps easier to follow and "bookkeep," has seen its day and nrore
than outlived iE usefulness. So I think most preparers and users of financial statements
are sold on the enhanced relevancy point. However, I am not convinced on the topic of
reliability of financial reporting using more fair value accounting methods. The increased
subjectivity and estimation process that underpins all aspects of fair value accounting calls
the "reliability" of such information into question for me. Having said that, t do believe
that fair value has a far befter chance of providing more reliable information in most cases
than the old historical cost model.
Look at the last four Statements on Financial Accounting Standards (SFAS) issued by the
Financial Accounting Standards Board (FASB), which all require some form of fair value
reporting (i.e., SFAS 155, Accounting for Certain Hybrid Financial Instruments, SFAS 155,
Accounting for Servicing of Financial Asseb, SFAS 157, Fair Value Measurements, and
SFAS 158, Employers' Accounting for Defined Benefit Pension and Other Postretirement
Pf ans, an amendment of FASB Satemens No. 87,88,106, and 132(R), and SFAS 159, The
Fair Value Option for Financial Assets and Financial Liabilities, Including an amendment
of FASB Statement No. 1 15). lt is very clear from these standards that the "fair value train'
has left the station, has accelerated quickly, and in my view will not to return anytime
soon.

Q: Similarly, what is your view on whether employing fair value iccounting measures
increases or decreases the complexity and "quality/ of financial reporting?
A: While l, and probably most other practitioners, think that today's accounting couldn't
;

possibly get more complex, I believe the trend to recording more asseg liabilities, and
the associated transactions at fair value will prove us wrong. Fair value accounting does
not ensure, in my eyes, enhanced "quality" in financial reporting. fu has always been
the case, the quality of the financial statemenb still hinges on the soundness of internal
controls over financial reporting and the thoroughness and technical competency of the
Preparers. I have heard many proponents of fair value accounting thought that it is the
end of earnings managetnent (and, therefore, improved quality of earnings) because there
would be more of a focus on the assets and liabilities on the balance sheet than on the
income statement. I would suggest that the inherent subjectivity and complexity of fair
value valuation techniques makes it ripefor manipulation, if that is theobjective. Of course,
the astute reader of the financial statemenb could detect some cases of manipulation if
there is robust disclosure of the assumptions underlying the
&i

valuations.

CHAPTER

Measurement theory

143

Q: During September 2006 the FASB

A:

issued SrAS t 57, 'Fair Valuc Masurements." Please

comment on what impact this will have on the curent accounting standards that rquire
fair value measurernents. Does this standard add to the complexity? Does it enhance the
consirtency of hir value reportintl
In my view SFAS 157 was long overdue. With numerous existing accounting standards
requiring varying degrees offair va lue estimates, SFAS I57 clearly provides some consistent
"how to" requirements for calculating fair value. However, it is likely to add some
complexity in financial reportin& as well as audit risk, associated with such reporting anci
disclosures. Moreover, based upon the assumptions within the "inputs' used to make the
fair value calculations, the Financial Accounting Standards Board established a hierarchy
of fair value measurernents for financial disclosure, referred to as Levels 1, 2 and 3. This
hierarchy, which is explained thoroughly in some of the articles presented in this iournal
issue, is intended to convey information about the nature of the inputs, with Level 1 being
the "most reliable," which includes inputs that are only quoted prices in active markets,
vercus Level 3, which consists of unobservable inputs lthat may reflect the Company's
own assumptionsl.
Of course, this new standard willrequire thal entities expend more e{ort than in the pasl
as they are{orced to use more complex valuation models and calculations that are claimed
to produce more reliable valuation estimates. Who is to make the final decision as to what
valuation technique is mo6t appropriate in each situation, especially when economists
have been arguing about these issues for years? | also foresee nrany instances where there
will be a lively debate about what the most appropriate 'hienrchical level" a particular
valuation estimate should be classified as, as Level 3 valuation requires more detriled and
costly reconciliations and rcll-fonryard disclosures for all activities that occurred during
the period. Nonetheless, this debate is healthy as it should improve the tiansparency
of fair value estimates to the astute reader and analyzer of the financial statements. The
benefit of having one set of guidelines for calculating fair value measures codifid within
one accounting standard will hopefully improve consistency and transparency which may
outweigh the costs associated with the complexity of the calculations. The disclosures
within the financial statements should be a good step in allowing investors to understand
the inputs and assumptions used by a Company to determine fair value and in turn, make
their own judgments on the reliability and relevance of the information within the linancial
statements.

Q: How does the greater rse of fair value accounting measwements affect the audit
procedurc of the independent auditor?

A:

The impact on the independent audit has been relatirrely extensive, panicularly in auditing
the more complex fair value calculations using valuation modell for example as required
by SFAS 123(R), Share-Based Payment, or in situations (which are becoming more and
more common) where a company uses thkd-party specialists to determine a fair value
measurement. ln short, in these situations the subjectivity and complexity has had a direct
impact on audit risk and the coct of the audit.
At KPMC, timely, exteosive and on-going traininS programs for its professionals
worldwide are developed and delirared on each new accounting and audit standard.
Additionally, our engagement teams are typically armed with a list of procedures and
steps they use to effectively and efficiently audit complex accounting areas, such as fair
value measures, refend to as audit program guides. Our firm also specifically trains
experienced audit and advisory professionals to assist engagement teams in working
through more difficult accounting standards, such as SFAS 157.
More often than no| our core audit engatenent teams are assisted by valuation
specialiss, actuarig, financial derivative resources or share-bosed payment specialists.
These profesionals, most of whom are specially-trained audit panners and senior
manatert principally assist the auditoE in assessing the qualifications of third-pany
specialists engagd by a client to calculate fair value measurements and disclosures

.and determining the reasonablenes of the assumptions and methodologies used in

such measurements. Of course, where valuations and estimates are used to determine

114

PART2 Theory and accounting pracrice

'I

fair values, there are numerous other audit procedures that are employed. These include
documentation and testing of management's process over calculating fair value measures,
including assumption development, as well as testing completeness, accuracy, and
relevancy of the underlying daa used in the valuations and other calculations.
Q: How could accounting programs appropriately prepare students for the fair value
accounting measurements they will inevitably deal with when they enter the business
world?
A: There is no doubt that the movement toward the increased use of fair value accounting
and estimation methods requires additional skill seu in accountants, auditors, analysts
and financial reporting specialists, than were previously required. I would argue that
the enhanced fair value requirements may significantly increase the workloads of these
individuals, given the complexity and the need for more continuous updating of fair
values estimates than before. Unfortunately, this comes at a time of already stretched
resources of both corporate accounting and financial reporting staffs and external auditors.

Accounting programs are going to have to ensure that there is a focus on teaching the
next generation of CPfu valuation techniques and in-depth financial statement analysis
tools. Additionally, higher education institutions may need to consider providing more
opportunities for students to specialize in areas of economics, finance, and statistics lin
addition to accounting degreesl, to ensure that today's students are adequately prepared
to ackle tomorrou/s challenges.
Source: Revrew ol Eusiness, pp. 6-9. College of Business Administrution at 5t. lohn'i University. New
York. Copyright oi Revier,r,o/ Busrness is the property of 5t lohn's Universily Coliege .

=}@-.

MEASUREMENT IN ACCOUNTING
in accounting hlls into the category of derived measuremnt for both
capital and profit. Accounting profir is now derived, under international acgounling
standardg from the change in capital over the period from all actMties including
increases and decreases in the fair value of net assets enduding transactions with
owners. Capital is derived from the net of 'fair rnlue' measure of assets and liabilities.
That means we have to measure the vdue of opening capital, the amount of income
Measurement

received, the amount of capital usage, and the change in the fair value of net assets. The
increase in capiul over the period will then measure the amount of profit from various

of
will then

sourcelr including operations and remeasurements (after adiusting for the infusion

new capital or payment of dividends). Ttre restated fair value of net assets
constitute opening capital in the nent period'

Contrast this measurement approadr with the approadr taken before the introduction
of intemational accounting standards. Revenue received was matdled against net assdts
used up in a period and if income was greater than net capital usage (or expenses),
then we had an increase in capital. Profit was not eamed until initial opening historical
cost capital was maintained and profit realised. That is, capitd was always stated at
historical costs and dranges in net assets were not considered as pro6t. Hence, we cln
see that derived profit depends very mudr on how we measire opening capital and how
we measure e:rpense and capital allocation. We can also see that the concept of capiul
valuation in accounting has errolved over time with the result Orat we have sareral
capital maintenance measurements and profit concepts. A brief historical overview will

illustrate this point.

ln the first thousand years AD, the economic structure

was rePresented by
was to count and
purpose.of
accounting
decentralised, self-contained fiefdoms. Ttre
safegqard the assets of the ste'lvard using single-entry accounting. Under this qrstem,
capital was measured as the stock of land, animals and agricultural produce with
CHAPTER
i.ii.
|L

Measurement

theory 145' ,.

the purpose of producing output (income) for sustenance. Capiul was not usually
rneaflred in financid terms but simply counted and itemised.
After the susades to the Holy Land in the eleventh century, the opening up of the
Middle Eastem and Asian uade routes created a demand for tradeable goods (sillc
spices, carpets).Ihe ltalian tnding cities played a maior role transporring cusadrs to
the Holy land and rauming with goods. This aaivity created a requirement for venture
apital. Profrt was based on the retums from (usually) single retum voyages, which
were financed by ventwe panner and calculared after retuming original apital. Thus,
ending capial was measured as the acormulation of wealth from individual vmtures
plus original capital. From a venture shareholder,s point of view, pro6t reprsmtd an
increase in wealth. Moreover, the use of the Anbic numbering systm together vrith
'.hc conacpt cf :Etu:nable capira! led ro rle er:elution of deuble-en'ry ec.o!nt!r,. '!'!.ris
system was used widely by Iulian merdrane from the twelfth to rhe sixteenth cmturies
and was first documented by Luca.Pacioli as the'system ofVenicd in 1494.
The eighteenth century in England saw the development ofjoint stock companies with
limited liabiliry, a sepante management class, and rransferability of shares. A number
of these companies were declared bankupL resulting in large losses to ceditoo which
in tum, led ro the introduction of rhe 1644 ,oint Stock Companie Regulation and
Registntion Acr. This Aa emphasised creditor protecrion and conservative accounting
rraluations. Thus, the definition of derived capiul moved towards ,creditor capiut,and
resulted in an acceptance of the lower of cost and market yalue rule as a meastlement
principle. In the nineteenth cenrury another concept of capiul emerged following
the railway orpansion in the United Stat6. This concept of capital revolved around
maintaining intact the stock of going concem asser (railroad ascets such as mgines,
coaches and tnck) so as to continue the ability of railroSds to provide the same level
ofuanspon services. This resulted in the concept of depreciation as a method to main
funds (capital) in order to replace asse6, and the going concem concept of capital
maintenance.

To this point in history there was litde developed theory of capital maintenance and

profi! only a collection of vague conceps. However, in l94O paton and Littletonr6

produced the 6rst defnitive srltement on the concepts ofcapital and profir They defined
as being derived fiom the matching or allocation ofhistorical cosB against revenue
eamed. Profit measurement was seeo as the major foors ofaccounting wirh rhe derived
balance sheet simply the repository of all ya-to-be-allocated hfutorical costs. Hence, rhe
balance sheet was not seen as a measurement of the net market value (or fair value) of a
busine*s. The concepts and principles of the paton and Littleton sysrcm formed the basis
ofthe conventional historical cost accounting system which was the dominant system
before the introduction of intemationat accounting standar& in 2005.

proft

The normative period of rhe l96os saw a number of challmges to rhe historical cost
principle ofvaluation and hence capital maintenance. Gritics deducrively argued rhat the
valuation of firms based on outdated historical costs was not all rhat useful for economic
decirion making and the derived profit did not measure contemporary resource usage.
They developed seranl capital maintenance and profit sptems bar.d on maintaining
intact opening capital adjusted for genenl and specific inflation. Thus, profit was derived
after maintaining intact some concept of 'market priced, capital, and viewed as a real
increase in purdrasing power or an ability to maintain the supply ofgoods and services.
There was robust debate about whidl was the 'dominant, profit measuremem system,
but the debate was never setded and lapsed somo,vhar in the litenture. Ttrese debates
can be considered the forerunner of the 'fair value, approach ro derived accounting
meatulement.

|
,

.,
,..- . ..1Li

PART

2 Theory and accounting practice

Richard Macve in his submission adopts a different, albeit equally critical, perspective.
An extnct from his conclusion is followed by an earlier summary paragraph outlining
specifics:
Even though it [FA58] a.gues that 'conceptually, fair value is a market-based measurement
that is not affected by facon specific to a particular entity' (para C2), this ED necessarily
ackno,r,ledge frat drere may be different market prices available to difierent enterprises, and
correspondingly different measures of fah value. lt proposec to resolve this divergence by
requiring that entities choose as their reference market the most advantageqls market to which
they have immediate accers (para C45). This issue is discussed funhei below. Howerrer, the
more fundamental issue of 'entr/ ve6us 'exif prices is left undiscused and unresolved (and
the definition of fair value is therefore itself necessarily left ambiguous) primarily because there
is no discussion of the underlying conceptual framework of valuation which has been fully
explored elsewhere in the academic and professibnal literature using the concept of deprival
value'for assts (and the related concept of'relief value for liabilities) (e.g., gaxte., 1925,
Chapter 12; AARF, 1998; ASB, 1999). lt is not clear whether the Board has come ro a view
on these conceptual isues or whether they are still to be addressed in the conceptual phase
of the pmiect. At present the ED gives the impression that the Board is merely avoiding these
issues
bul until they are addressed and explained the implications of the definition of fair
value and related requiremnts of this proposed standard are necessarily left incomplete and
essentially unclear. (For funher discussion see Horton and Macve 2000.1 [R. Macve, FASB
Submission f6, p. 2l

Brief discussion of those two events reveals the difficult path ahead for the IASB countering
such will-o'-thewisp issues to achievea convergence regime that produces financial reporting
comparability acroas countries.
Jou.ce.' Excerpts from.4bJcu.r, r,ol. 41, no. 2,.2005.

Questions
l. What do you think the author mean by a principles- versus a rules-based system of
accounting?

2. What are the measurement problems they allude to?

3. what does Macve mean by the undrlying conceptual
see

4, Do you think

148
I

PART

framework? What problems does he

with a 'fair value' apPtoach?

there is any difference between 'measurs' and 'values'l

For some firms it is argued that frir value accounting fundammtally changes the
focus of risk management. That is, firms will decrease their hedging aclivity bcaus
they are wonied about the accounting impact of profis under IAS 39/AASB 139. One
other consequence is that a company's pension fund now shows up as a liability
Benefts) and this may need to be
on rhe balance sheet (lAS I9/AASB f19 E
hedged. The derivatives a 6rm uses to hedg this liability might depend on whether the
pension Echeme is in surplus or deficit Thus, intemational accounting standards may
reduce hedging activity if it ruults in an increased volatility in income while increasing
hedging and risk management for pension liabilities.
The IASB also makes a uade-off betrreen reliable measures and relevant measures.
Sometimes that judgement is made by the [ASB. For mmpls in IAS 39/AASB 139,
for availablefor-sale securitis the IASB daermines that fair value (selling price) must
be used instead of historical cost. However, for IAS 41/AASB l4l Agriatlture, although
fair value is required where possible the Board stats that it may be difficult to obtain
reliable measures of fair value. In rhis case the reliability e<ception allows preparers to
make a trade-off benveen fair value and reliability when measuring value.

2 Theory and accounting practice

npw

:::

Consequently, we were left with a number of accounting mea,surement Erstems.

These different perspectives reflect various boundaries of accounting and a lack of
agreement on measurement principles, but with the historical cost allocation system

a:

as the conventional and dominant model. Added

to this were a number of academic

accounting Papen that suggested the value rela'ance of conventional profit had
significantly declined over time, but balance shea items and intangible assets had
become more important.lT More recently, the Intemationa! Accounting Standards
Board (IASB) has taken the vieru that globalisation of business supporrs the need for
one set of accounting standards to be used throughout the world in order to produce
comparable fi nancial information

ii

This has resulted in trn'o notable developments in intemational accountingstandardsls

as signalled through accounting standards sudr as IAs 39/MsB 139 Financinl
tnstruments: Recognrnon and. Measureffient and the LNsB/FAss ioint proiect on reponing
financial performancetl
(1) that profit measurement and rerrenue recognition

should be linked to timely recognition, and (2) that the'fair value'approach should
be adopted as a working measurement principle. firus, from 2005 we see the use (in
part) of a measurement principle that focuses on the drange in the value of assets and
liabilities rather than the completion of an eamingp process. In short, this means that
changes in the fair value of assets and liabilities are recogrised immediately they occur
and reported as a component of income. Furthermorg the focus has shifted towards
a valuation concepL with the bdance sheet the maior repository of value-relevant
information, and the main users of accounting information stated to be shareholders
and investors. At no stage has the principle of capital maintenance been orplicitly
discussed. These measurement concepts are not without controversy. rheory in
action 5.1 comments on fair value measurement within IASB and Fr{.SB standards and
raises a number of concems surrounding fair rnlue accounting and whether it
has an
underlying measurement principle.

..?

l.l
Itlr(!R\
\(-ilo'r

Fair value measurement

'True and fair'and'fair value,

and legal will-o'-the-wisps

accounting

by Craeme Dean and Frank Clarke

L!oj*Y" (IFRS) regime to press as the convergence to an international financial reporting

standards "^t-!!3*t.g*:
begins. More than thirty years since the lengthy gestation of thi
International Accounting Standards (lAs) committee, application

ind inforcement

of

international standards under the guidance of the International Accounting Standards Board
(IASB) is nigh ...
Within.this. seftie
,
$g suggestion by many commentators (including the lASg in its public
documents) that the IFRS regime is principles- rather than rutes-basid is contestable. This
matter has been discussed in several rec ent Abacus editorials and is further examined here.
Whereas those promoting the principles rather than rules mantra have failed to specify what
those principles are, the general tenor of their comments gives reason to imaginethat such a
distinction is underpinned by the qualitative criterion, 'trie and fair view, (oi its equivalent)
legal or professional override. Frequent recourse to 'fair value' in many IFRS adds further
to a principled approach- to things commercial. lt will be shown that both settings,
.support
however, illustrate the contestable nature of what is meant by those preferring principll.
Those seftings represent potentiat difficuhies in achieving .onu"rgun." of an eiective iFRs
regime

...

fijc

CHAPTER

Measurement

theory 147

Capital or income?
Dr

Nadine Fry and Dr David Bence are senior lecturers at the Eristol Business School

Despite the established link between asset and liability measurement and capital maintenance,
the latter represents a second order issue for the IASB, say Nadine Fry and David Bence.
In May 2005, the lnternational Accounting Standards Board (IASB) and the US Financial
Accounting Standards Board (FASB) launched a joint project to develop a common conceptual
framework that could be used to develop new and revised lnternational Accounting Standards
(lASs), As part of the project, much time will be spent considering the measurement of business
income.
The 'debit' side of the problem has received a substantial amount of attention in recent
years with the IASB's drive towards fair value as a solution to accounting measurement
problems. Yetdespite the established link between asset and liability measurement and capital
maintenance, the latter represents a second order issue for the IASB. 5o if a futures contract
has a fair value of f l0m but an historical cost ol zero, should one credit income or capital?
lnterestingly, although the IASB's Framework claims that iinancial statements should
enable users to determine distributable profits and dividends, the IASB argues that such
issues constitute part of the legal framework of individual countries for which there is little
i

nternational convergence.

Regrettable
However, even if one accepts that there

will always be different distributable profit

rules

around the world, the IASB's failure to decide on a capiAl maintenance concept is regrettable
as users have no idea as to whether total gains represent income or capital and are therefore
unable to identify a meaningful 'bottom line'.
The fASB does not appear willing to Ackle capital maintenance issues. IAS 1, Presentation
of Financial Statements, permits companies to produce either a 'statement of recognised
gains and losses' or a 'statement of changes in equity', but both ignore capital maintenance.
Although it may be argued that sophisticated users can make up their own minds as to the

nature of total gains, this decision is not in line with the FASB/IASB conceptual framework
project, which aims to deal explicitly with the issue of capital maintenance.
The approach currently adopted by lASs generally provides a measurement of incorne
based on the nominal unit of currency, a mixed system for measuring assets and liabilities
and money capital maintenance (see tbr examplg IAS 19, Employee Eenefits; IAS 2'1, The
tffec9 of thaiges in Foreign Exchange Rates, IAS 39, Financial lnstruments: Recognition and
Measurement, and IAS 40, lnvestment Propertyl.

Inconsistent
Surely a mixed measurement system and money capital maintenance do not 8o together
well? However, the concept of money capital maintenance is not consistently applied
across international standards. For example, IAS 29, financial Reprting in Hyperinflationary
Economies, has the effect of real financial capital maintenance, whereas IAS 16, Property,
Plantand Equipment,and IAS 38, tnangibleAssets, apply physical capital maintenance.
It seems fair to conclude that even if the lAsB continues to apply capital maintenance on
a line-by-line (or element-by-element) basis, an additional financial accounting element, a'capital maintenance adjustmenf, should be included below'total gains' in the 'statement of
recognised income and expense'.
Thl nSB could provide guidance on the items that would make up the capital maintenance
adjustment, presumably iniluding changes in the unit of currency, unrealised gains, realised
gains where'replacement is required, and so on. Such a financial accounting element would
inabte users todetermine the proportion of total gains represented by capital and income and
would help to alert users to th; fact that some gains may reverse because they are unrealised.
S ou

rce: Ac co.t n tanc y, April 2rJ07, p.

1, www.accou nlancymaga

ne'com'

CHAPTER

Measurement theorY

Questions

Why is the measurement distinction between capital and income imponant?

2. What do you think is real financial capital maintenance and wh;t is physical capital

3.
4.

maintenance?
ls the increase in the value of a house you live in income or capital? provide explanations.
Does your answer to question 3 change if your house is an investment property? Cive
rcasons tor votrr answer.

The

ioint proiea on

Financial Statement presentation (previousty

Reponing Comprehensive Income or performance Reponing) highlights the IASB,s
thinking on income and i*ser meixrurcmenL pa iculady rhe ipprication of fair value
FASB/IASB

mearurcment. Some agreed concepa include the following.ro

1. Accounting information should be aimed at decision makers making economic

decisions about the entity.

2. Eiitities should present

a siri:gle statement ofall recognisid incorne and orpense items

component of a complete sa of financial staternents.
3. The staterient should be all-indusive:
(a) k should indude the effecrs of all dranges in net assets and riabiliries during rhe
period, other than transactions with owners.
as a

(b)

liabitities shourd be valued at fair value which pruumes market prices

but substitutes such as discounted future cash flows, depreciated marker prices
or asset-pricing models can be used in the absence ofa liquid market.
(c) Income determination should be split between prcfit befol remeasurement
and
Assets and

remeasurement effects.

4. All income and oEenses shoutd be categorised and displayed in

(a) enhances users' undershnding of achieved perfornanie
(b) assiss in forming expettations of future performance.
5. Profirs should not be based on a notion of realisation.
6. lhe focrs should be on

way rhat

(a) greater transparency

(b) useful infomation to investon and relevance of data
for decision making
concept of reliability has been replaced by represen,",ionit
f"i hfulnor.
_(c) .the
under this approach the income st,'t"ment wourd become the
residual between
closing net assets, rather than the balance sheet becoming
:t_.lli.g,
"T of1r.*,:nd
the residual
unallocated cosB after the matching process, which is the case under
historical cost measuremenL The concepts srated ab;v; give an
inaication oroe Boards,
thinking about financial statement presmtation and m;surement
issu6.

r-.

MEASUREMENT ISSUES FoR AUDITORs

Serreral issues are created

t50

PARI2

for auditprs by the shift in focus for profit measurement from

matching revenues and openses to assessing the change in u. rair
uano

when profit is determined by matching reuenue ant

Jn"i"rr*.

er(pnse transactions

for rhe

period the auditor can (oncentrate on gathering eridence

tf,at those transactions have
been handled appropriately by the clienCs accounting system.
Howwer, when profit is
derived from changes in fair values more difficult queitions
a*"rot rrt"
gathering evidence dn managementt estimates
"ujiior'".*a
For orarnpre, one aspect of measuring profit by assessing
changes in fair value of net
assets is addressed by the accounting standard IAS 36/A458
tgo. ihis standard requires
a dedine in an asset varue to be recdgnised as an impairment
ross. Management ofthe
entity is required to assess at rEioning date whether there is any
indiJ on that an
Theory and accountinB practice

.l

;TF:T
:,

:i,ii:
.1';

asset may be impaired. lf any such indication odsts, management shall estimate the
recoverable amount of the asset. If the recoverable arnount of an asset is less than its
carrytng amounL the carrying amount of the asset shall be reduced to its recoverable
amount. That reduction is an impairment loss. The impairment loss shall be rgcognised

immediately in profit in most cases.

The intemational auditing standards guidance for auditing impairment losses and
other fair value estimates is contained in ISA 540.21 Auditors are required to gather
evidence to iudge if management has followed the accounting standard appropriately
and if the amount recognised as an impairment los is reasonable. To do this, the auditor

muit detennine whether management has chosen appropriate and reasonable rnluation
methods and assumptions. If the accounting standards do not prescribe the valuation
method forthe particularassets and liabilities beingconsidere4 theauditorcould accept
any reasonable valuation method. As orplained in deuil in cdse study 5.1, there are at
least twelve methods of valuing intangibles and brands from which management could
choose. This means that it is diffcult for an auditor to disagree rvith managemenr's
choice of a particular valuation method that is being used by other entities. The auditor
must gather evidence that the method is applied consistently, so that managers do not
pick and choose methods from year to year depending on their desired profit result.
Auditors must also assess whether the asset or liability values are properly determined
from management's significant assumptions, the valuation model and the relevant
underlying data. Such data would include the interest rates used for discounting cash
flows, market values used by comparison companies, royalty data, and so on.
Overall, given the odstence of different reasonable valuation methods and poisible
assumptions, it is possible for several different but reasonable amounts to be recognised
by management for impairment losses. Ttrese different amounts would therefore be
acceptable to the auditors if the audit evidence suggests that management has applied
the valuation models correctly and used appropriate data. In these circumsunces; it is
posible that auditors face pressure from managen to agree with their valuation choices
or else lose the audit to another auditor who is more agreeable.
In addition to the issues related to the use of fair values and associated issues, auditors
also face problems caused by rnriability in the level of reliability and accuracy of
measurement of historical costs. For otample, standard costing manufacturing systems
are based on historical costs of various inputs, assumptions about processing volumes
and methods, and issues sunounding the assignment of overhead costs between
products, processes, and departments. All these factors affect the costs of inventory on
hand at the end of the period and goods sold during the period. ln this conto0 the
auditor needs to test the reasonableness of the procedures adopted in dweloping the
standards from the engineering specifications. This includes gathering evidence about
the reasonableness of the underllng assumptions and consistent use of the data.
The inventory cost per unit will appear to be very precise, but changes in operating
conditions could produce significant variances and render the underlying assumptions
for cost allocation to be invalid.

CHAPTER

Measurement lheory

15r

Eil=

The importance of measurement

Measurement involves the formal linking of numbers to some property or event by means
of semantic rules. Ttre semantic rules in accounting are represented by transactions in
markets, and the allocation of the use of capital resources against incoming revenue over
the period.

The nominal, ordinal, intervaland ratio scales of measurement

The rules used to assign numbers are determined according to four scales: nominal, ordinal,
intenal and ratio. ln accounting we use the ratio scale to measure the financial attributes of
ProfiL assets and liabilities. However, we may also applythe ordinal scale to rank investment
proiects or the profitability of firms, or the interval scale in standard cost accounting.

The permissible operations of scales

Invarlance of a scale means thaL regardless ofthe measure used, the measuremenr s)'srem
will provide the same general form of the rrariables and the decision maker will make the
same'decisions. This is not the case in accounting as different systems are variant to each
other. Income measured under one system witl lead to different decisions from income
measured under another s)'stem. The systems won't provide the same information.
The difference between fundamental, derived and fiat measurements
There are three different t'?es of measurement. Fundamental measurement is when
nurnbers, which do not depend on other properties, can be assigned by reference to
natural laws. In accounting there is considerable debate over the nature of fundamental
value. Derived measurement depends on the previous measurement of two or more
other quantities, for orample, the determination of income and expenses before profit
is determined. Fiat measurements are those that relate numbers to properties of objects
or events on the basis of arbiuary definitions. AII measurement invotves erors and for
many measurements the true value is not known.

What is meant by reliability and accuracy in measurement

Reliability refers to the proven consistency of the measure, that is, the same (or similar)
measure would be reproduced by different observers. Accurate measure refers to the
representation of the fundamental value of the property being measured. Measurement
theory also teaches us that rnany of our measurements in accounting are on a ratio scale,
which is the most informative scale, but they have the weakest theoretical foundation
bbcause they are fiat measurements. Greater confidence in such measurements can be
attained if there is theoretical or empirical evidence ro support the relationships of the
properties or the need for such theories. However, all accounting measures contain

:?'

t--

PART

Theory and accounting practice

gt

t;;-l
a-,1

:.-

measurement risk and there is considerable debate in accounting as to what constirutes

fair ralue and what are the acceptable disuibution ranges. An understanding of the
concepts of risk and uncertainty help us to understand the valuation problem.

Measurement in accounting
The two fundamental measures in accounting are capital and profit and they are both
derived measures. Capital is dedved from uansactions and revaluations that occur in
financial markets, and profit can be derived from the matchingof expenses with revenue
or the change in capital over the period. Capiul can be defined and derived in various
ways, induding historical cosl operating financial, or'fairvalue'. History shows us that
the concepts of capital and profit have changed and evolved over time so that there are a
numberofconcepts offundamental measurement. Morerecently, intemational financial
reporting standards have made greater use of the concept of 'fair value'. A number of
cornmentatoF argue that this concept diverges from allocation principles to a valuation
approadr, whidr will differ according to the circumstances and subjective interpretations.
These changes focus more on balance sheet valuation, moving accounting away from a
simple profit allocation measurement system and placing more emphasis on relevance
for commercial reality and investor decision making rather than reliability.
Measurement issues for auditors
Ttre oristence of alternative valuation methods for some assets creates issues for auditors.
There could be many different asset values that are acceptable to the auditor if the
valuation method is applied appropriately and consistently, reasonable assumptions
are used, and the data used to generate the vduations are valid. The auditor could face
pressure from managers to accept their valuation or the entity will seek another auditor.
There are also isues in auditing historical costs, sudr as standard inventory costs, where
the costs are precisely stated, but based on'assumptions about engineering processes that
are influenced by dranging conditions.

Questions
r. Tedrnically, what db we mean when we say X was measured'?
2. How is a scale related to the process of measurement?
3. Describe the following scales: nominal, ordinal, interval, and ratio. Give an
orample of each. Wht t, scales are applied in accounting and where?
4. Determine whether the following statemenB are conect and state why
(a) The historical cost of inventory is $60 000 at )'ear-end' when it is converted
to constant end-of-year dollars by multiplying by r 10/100.to get $66 000, the
$66 000 is still the historical cost of the inventory'
(b) Last month the quantity variance was daermined to be $12 000 favourable,
and this month it is $24 000 favourable: therefore the efficient use of materials
has been doubled.
basis of saving income ta:r, C-omPany X ascertained that the
diminishing-balance depreciation method is bener than the stnaight-line

(c) On the

method. By using the diminishing-balance method for depreciation rather

than the straight-line method, Company X saved $tO OOO in income ta,t this
year; therefore, the fogner method is 10 000 times better than the latter.
(d) On the basis of the amount of assets, we can say that C,ompany X is twice as
large as C,ompany Y, because ie totd assets amount to $ I 000 000 compared

with $500000 forY.

5. Describe the following t1ryes of measurement fundamental, derived, fiat. In what
sense are fiat measurements 'weak

What type of measurernent is inventory costing?

CHAPTER
t

Measurenrent theory

153

rli

rl

6. \tfhat are the sources of eror in measurement?

z. nrphin whether the following shtements are facts:
(a) Canbena is 320 kilometres from Sydney.
(b) Depreciation e,(pense for Ikmbah Pty Ltd for 2008 was $ I 294 000. (This is the
amount reported on the income statement.)
(c) Smoking leads to lung cancer.
(d) sales r:'/enue fcrTelex Ltc for 2009 r.ras $2 900 c00. (This is the amount
reported on the income statement.)
(e) Equipment (net of accumulated depreciation of $400 000) for McNair Ltd for
2008 was worth

$1

800 000. (Ttris is the amount reported on the statement

of

financial positon.)
8. What is the difference between accuracy and reliability in measurement? How are
these notions related to the testing of a rheory?
9. Discuss whether accounting measurement is fiat or fundamental. Can accounting
numbtirs evCr bC relatdd to-ftindamental Values? If so, rrihat alesomL possible
measuremens that can be used?
10. Discuss how recent accounting standards using'fair value, accounting such as

LAs 39 and IAS 4l have moved away from fundamental measurement.

1r. Why will international financial reporting standards induce an increased demand

for risk management techniques?

12. How will the use of fairvalues affect the role of auditors and the
audit function?
Do you think it will affect the training of accounting students? How
so?
13. what would be included in overhead costs assigned to irruentory?
How many
wa)ts Gtn you think of assigning these costs when there
are multiple products and
depanments? What are the audit implications?

Additional readings
K, picker, & loftus, L Crark, K & wise v 2009, Appwng intqnational
repnting standards,2nd edn, Brisbane lohn wiley & sons.
_financial
Charles, E 2}O2,'Mi:red ba{, Ausnalian CpA, May, pp. Z6_SO.

Alfredson,

K ko,

Churchman,

cw, &

Ratoosh,

P (eds) 1959, Measurements:

New York: Iohn Wiley.

itefinitions

anil

theories.

Institute of chartered Accountants in England and wales (rcAElv)

2006, Infmmation
for better marheu: measurcment in financial repurting, icetw'www.icaew.co.uk/
bettermarkets.

Karn, V 1973, 'ludgements and the scientific uend in accounting',

Iournal of Accountancy,
February.

following article discusses 12 differe-nt_ways to value inhngibles and brands

using
values'. The article highlights the diffictity involved
*liability anj
accuracy and in providing relevant informaiion.
T'he

'hir

iild6g

12 ways to value intangibles and brands

l.

Net present value The mainstream discounted cash method, used

by the Big Five
and Interbrand. This method look at the proportion of business pr#s
driv-en by
brand and the relative security of branded profit stre"ms.
2. capital asset pricing model The main mehod used by Brand Finance,
which assigns
a brand to the intangible. This is interchangeable witir the
cApM,s di'scount rate.

154

PART

Theory and accountlng practice

rt:

lil
I

I
;
J
!

'

3.

4.

Cost of creation This attempts to calculate the cost of building from scratch. Some
say this technique is unworkable and unrealistic.
Market-based comparisons This is literally what it says. A comparison is made to a
similar brand in the market. ln practice difficult to achieve. One only need compare

two similar tasting drinks, Coke and Pepsi, and their respective brand values of
US$68.95bn and U5$6.2 I bn respectively.

5. Royalty relief method lf a brand is licensed from a third party, a royalty is paid.
Here future'sales are estimated and applied to a royalty rate to arrive at future brand
value. The cashflow is discounted back to a net present value or brand value.
6. Relative value This looks at valuing the progress a company has made towards
meeting target. For example: have 80 per cent of employees been with the customer
in some meaningfulway?
7. Balanced scorecard Supplements traditional financial nreasures with three
additionalperspectives such as customers, internal business processes, and learning/

growth.

8. Competency models This would typically calculate the market value of the
behaviours of a company's most successful employees.

9. Benchmarking Involves identirying companies that are recognised leaders in

leveraging their intellectual assets, determining how well they score on relevant
criteri4 and then comparing your own company's performance against them.
10. Business worth This approach centres on three questions. What would happen
if the information we now use disappeared altogether? What would happen if
we doubled the amount of key information available? How does the value of this
information change after a day, a week, a year?

11. Business process auditing This measures how information increases value for

specific business process, such as accountin& production or.marketing.

12. (nowledge bank This turns around traditional accounting by saying that capital
spending is an expense rather than an asset. ln contrast, salaries are treated as an
asset rather than an expense as they create future cashflows.
Sr,urr:e.. ,tustrafi;tn CP,\, May

2002. p. 29 tusing ihe sources ol Montague lnstitul,e Reviclv,

lnterbr.rncl, Eranel Finance).

Questions

1. Evaluate the advantages and disadvantages of each suggested measurement

technique.

2. Make i recommendation as to which measurement technique should be used.

3. Would you consider using more than one measurement technique? Why or why
not?

The fotlowing article raises concerns about the impact of international accounting
standards on risk management strategies. ln particular, it highlights the impact of
tAs 39/AASB 139.

Risk management survey 2005: Take a risk. No chance

by EdwardThornton

How is IAS affecting corPorates

,For some clients it s lecreasing their hedging activity because they are worried about
the accounting impact' says Rlshid Hoosinally, head of client strategies and product
innovation at beuische Bank. 'But I think other clients have turned the corner'' Having

CHAPTER

Measurement thmrY

Ii;i

t5s

fi
r

been initially fearful of IAS 39, Hoosenally believes many corporates are determined
to implement strategies $at are in the long-term economic interests of their company,
even if it means bearing the pain of volatility in the short term .. .
This observation is certainly borne out in [Corpora te Finance'sl survey. While nearly
half of respondents say that accounting considerations were ,very important,, the vajt

'li

l;
'j

r,
ii
ir

majority (89.3670) of corporates surveyed say the ratio of their hedging done via options
and structurecj forwarcis haci not changed since IAS 39 becameln-issue. Corporates
t9 be in a quietly defiant mood: acknowledging the importance of accounting
:eem
issues, but refusing to be cowed.
On the other hand, says Russell Schofield 8ezer, managing director of the corporate
financial engineering group at JPMorgan, the issue of strictir iccounting regulatiohs has
fundamentally changed the focus olrisk management. Before, much-of corporate,s
time was spent worrying about the steepness of the dollar curve or the Federal
Reserve's next interest rates announcement. ,we don,t have many conversations
around those micro topics anymore,' he says. rAll the conversations *L h.uu with
our
corporates are fundamental in terms of "what's the impact on the business
now of risk
management?"'

ii''t

9n.

ri
'l

consquence of the stringent accounting regulations is that a company,s pension

fund now shows up as a.liability on the bala-nce-sheet. And, rile an/'otrr!,

ii"bility,
,: b:-l."dg* The derivatives a corporate uses to hedge thi's riabirity mighi
l,j1{r

:I
I

depend on whether the pension scheme is in surplus or deficit.

companies that have. large deficits are'thinking if ttr"y start to hedge
..
their
,h"y, effectivery rock themserves into tr,i r"r!" iiruiiiy tr,rt i'as rear
lll9ll
losses,'::l
says Deutsche Bank,s lRichl l-gsyrnsr.

]llf

n::

-$burcr:: E.rcerpts irom

grrrrrn"r*

Fin,tnc.t,,lsstre 2.11, April 2005.

Questions

l'

Wly would firms be concerned with increasing or reducing hedging activities under
different accounting systems?

2' what impact will international financial

statement?

reporting standards have on the income

3' wly

is the measurement and reporting of pension liabilities

and assets a concern for
risk management?

The trend toward fair value accounting

by

I Russell Madray, CpA

The Debate
profession so so
,:li:T:.::1,:I:gi:-::iT:1t$*ed. some in rheaccounting ;ffi#"rffii
lT"?.:":"-f":::1.:il"gl,statem;n*.alr"ri."rprJ"ii""#;'f
fact at tr,",.*"t u. ;; ;fi ffi ii'ff ;' ;'il:;
lff ll:;
3i:,y:::_",TI ll 1y ll"
th

jrfr{rT::1q;G;';rffi#;;;'ff";t
]ess .r,"i-an Robert
l[l?:i:lT:::T,:t':,':^::::::E:
*I,:tI: y..iy reformers, in.rrain!;l;;r
to making financiar"",1ilrilil'lf,li
sratements more
3::::I"-::.:::j,Tg Ty:l!".p"n:lge
j.;;;;;";;;;il;&='[';#";
jl'1".'l:i
rj:::g,".:'-::*:19:,::-':-:1
*lr::f :*::11'Ji^:r,:'-'.*!1.^1,:e.oril,;;;;;#:,#"i"';;;#;l
_

a, vg[EYg

t]tdt

ji:,:::f ::s*ofairvarue,"q;t;;;;;;;;.,ffifi i;insassers

?*:y:,:ITj:":
ca
jn=lf:.:il
:l ::,y;_'":t:ts 1 be very * bieai ;. r.i's ili'rc il' ilffi ;;;

less.retiable. rn facr, disputes..n

liabilities.

r55

PART

Theory and accounting practice

"ri*

ou"i

"",
d&;;:f#il"j':"J;il
#TT:

f-

-l
The crux of the fair value debate is this: Each side agrees that relevance and
reliability
are important, but fair value advocates emphasize relevance, while historical coJt

advocates place greater weight on reliability.

Relevance versus Reliabil iO,

The pertinent conceptual guidance for makingtracle-offs between relevance and reliability
is providel by FASB Concepts Statement No. 2, Qualra tve Characteristics
of Accounting

lt provides guidance for making standard-setting clecisions aimed at

producing information useful to investors and creditors. Concepts Siatement No.2
states:
The qualities tlrat distinguish "better" (nrore useful) intbrmation from ,inferior,' (less
useful)
infommtion are primarily the qualities of relevance and reliability ... The objective
of
accounting policy decisions is to produce accounting iniormation that is relevint
to the
lniormation'

purposes to be served and is reliable.

Critics of fair value generally believe that reliability shoulcl be the clominant
characteristic of financial statement nleasures. But the FRS'B has required greater
use of
fair value measurements in financial statements because it perceives that iniormation
as more relevant to investors and creditors than historicai cost information. ln
that
legard, the FASB has not accepted the view that reliability should outweigh relevance
tbr tinancial statement measures.
Some critics also interpret reliability as having a meaning that differs in at least
certain
respects irom how that term is defined in the FASB's Conceftual Framework.
Some critics
equate reliability with precision, and others view it principally in terms of verifiability.
However, Concepts Statement No. 2 defines reliability ai "the quality oi information
that assures that information is reasonably free from eror or bias and faiihfully represents
what it purPorts to represent." With respect to measures, it states that ,,[tlhe ieliability of
a measure lests on the faithfulness with which it represenB what it purports
to represent,
coupled with an assurance for the user, which comes through verification, that it has
that
reprcsentationa-l quality." Thus, the principal components of ieliability are representational
faithfulness and verifi ability.

Although there are reliability concerns associated with fair value measures,
particularly when such measures may not be able to be observed in active markets and
Sreater reliance must be placed on estimates of those measures, present-day financial
statements are re.plete with estimates that are viewed as being sufficiently reliable.
.. ..
lndeed. present-day
nrpsenl-dav measures
mercrrrac af
mrn.r assets
rccolc and
rnrl lir!.ili+;-.
Indeed,
of many
(and' ^u^--^-:-'^l-:=rr'rirj.iiir*i
liabilities t^^j
based on estimates, for example, thecollectibility of receivables,
useful lives of equipment, amounts and timing of future cash flows
likelihood of loss in tort or environmental litigation.
Even though the precision of calculated measures such as those in clepreciatidn
ac.counting isflot open to question since they can be calculated down to the penny,

the

reliability of those measures is open to question. precision, therefore, is not a component of.reliability under concepts statement No. 2. In fact, concepts statement No.
2
expressly states that reli.ability does not imply certainty or precision, and adds that
any
pretension to those qualities if they do not exist is a negation of reliability.

'

Rohert H. Hen's remark tn the Financial Exrrulives Intemational Current Financial

Reporting lssues Conference,
Ncw York Hilton Hrlcl, November 4, 2fi)2.

S-ource: Excerpts fnrnr 'The trr:ncl loward fair v.rlue accounling', lournal

Prolusionals, May 2008, pp.

of

Financr.r/

lfi-.I8.

5enice

Questions
l. Whgtyou think isthe fundamentalproblem with financial statements based upon
the historic cost measurement principle used under US CAAp ?
'. . . accounts must reflect economic reality, as a
core principle of measurement in accounting?
How would you measure economic reality?
What is reliability in accounting?

2. what do yog think of the principle

3.
4.

CHAPTER

Measurement

theory I't7

li
The following article discusses share option schemes in the light of the international
financial reporting standards.

IFRS
by

r'
I

put damper on share options schemes

Barney lopson

International financial reporting stanciarcis have triggered a sharp cirop in the popuiarity
of share option schemes, realising the tears of critics who said the rules would kill off a
vital recruitment tool.
Under IFRS, companies are required to deduct the cost of issuing options from
eatnings for the first time. This has led to sudden reductions in reported profit at some
companies.
As a result, stock options are falling out ot tavour in the remuneration ot l-l5h 100
chief executives. The proportion of incentive awards made up by options has dropped
to 21 per'cent this yearfrom 36 per cent, according to PwC.
Graham Ward-Thompson, partner at PwC, said: 'IFRS is creating a fundamental shift
in the design of stock-based incentive plans''
to
notably small technology groups
Options have enabled many comPanies
recruit elite and rank-and-tile staff who are attracted by the chance of earning far more

than cash salaries when share prices rise.

OpponenS of option expensing have warned that its effect on profits could make
the initruments unusable, leaving companies unable to attract good sAff and inhibiting
innovation.
PwC's research showed that granting actual shares had become a viable option.
'Performance share plans' now account for 68 per cent of FTSE 100 chief executive
incentives, up from 57 per cent.
Performance share plans must also be expensed under IFRS, but Mr Ward-Thompson
said they enabled companies to get the same'incentive and retentive effects' for a lower
accounting charge. 'Employees don't see as much value in options as they do in free
shares.'

Companies that have introduced performance share plans for senior management

or empioyees include BM, British Land, Carphone Warehouse, Emap and Legal &
General.
Stock option expensing is expected to exert a drag of about 13 per cent on the Pretax profit of LogiciCMG, an lT services group. Seamus Keating chief financial officer,
said'the compJny was considering changes to remuneration but would have done so
irrespective of IFRS.

to incentivise employees
'Siock options have probably been a less effective tool

recently because the stock market has been flat and the tech sector has gone through
some difficult years,' he said.
Aim,
James Wheaton, finance director of Business Systems Group, which is listed on
,l
said: don't think you should run the comPany on a change in accounting policy.

,lf options were the right way to find and retain people we would use them and

explain it to shareholders.'
Sourc<': Financi,zl limes, 1 | August 2005.
Questions

t,

How do share options schemes affdct the measure of income under international

financial reporting stqndards?

2. Why has this led to the reduction in stock-based packages for staff?
3. How is the form of structuring and reporting financial statements important to
(a) management?
(b) shareholders?

158

PART

2 Theory and accounting practice

Endnotes
l.

N Campbell, 'Symposium:
measurement and its imponance
for philosophy', Proceedings ofthe
Aristotelian Society, ilfe, p. tZe.

2.

SS Stevens,

'Mathematics,
measurement and psphophysics,,

in

SS Stevens

(ed.). Handbooh of

etcperintmnl psychologr, New york

Iohn Wiley, 195t, p. l.
3. R Sterling Theory of tlv nre.r:itrerneilt

ol enterprke income, Y:iansaf,: Kansas

tlnivenity, 1970, p. 10.
4. Stevens, op. cit., p. 2.
5. Srevens, op. cit., p. 23.
6. W Torgerson, Theory and netlrcds of
scrllfng, New York fohn Wiley, 1958,

p.17.
7. ibid., pp. 21, 30-1.
8. R tvlattessich, Accounting and
dnalytic.tl methotk, Homewood,

10. Nunnally, op. cit., pp. 16-17.

11. N Campbell, Physia, the elemenu,
Cambridge: Cambridge Univenity
Pres, 1920; N Campbell, An accouttt

of the principla of neasurement anil

cdathttiotrs, London : longmans,
Green, 1928.
12. W Hays, Quantificatiott in pq'chology,

New York Brools/Cole, t967, p.

lZ.

13. B Ellis, C.ancepu of meannammt,

Cambridge: Cambridge University
Press, 1966, p.53.
14. Torgerson, op. cit., p. 24.
15. SS Wilh, 'Some aspects of

quantification in science', in Harry

Woolf (ed.), Qil(rnfrFcurrlon, New
York Bobbs-Menill, 1961, p. 7.
16. W Paton and AC Littleron, Arr
introduction to corpoftrte dccounting

lll.:

Inrin" 1964, p.7t.

9. Stevens. op. cit., p. 25; lum
Nunnaf ly, Psychomenic rlreo4,, New
York: lvlcGraw-Hill, 1967, p. 15.

snndutb, Florida: Al{A, 1940.

I 7. See

Francis

and K Schipper, ,Have

financial statemenc lost'their

relerrance?', Iotnnal
Rese.rrcft,

of

Accounthry

Autumn 1999, pp. 319-52.

8. Intemational accounting standards

were adopted from 2005 in Ausualia
and the EU countries (forconsolidated
accounts oflisted companies) and
from 2007 in New&aland.
19. Deloitte Touche Tohmatsu lASplus
websire, IASB Agenda project;
Reponing Comprehensive Income
(Performance Reporti ng),
I

www.iasplus.com.
20. Financial Statement presentation,
www.iasb.org; see also project
Updates: Financial Statemenr
Presentation
foint proiect ofthe

IASB and Fl,S& www.fasb.org.

2I.

lntemational Standard on Auditing

540: Auditing trccounti,rs estirnates,
ittcludhry lhir value uccotmrhrg
estinuttes, and relueil ilkdosures.
The equivalent Australian Auditing
Srandards are ASA 54O Arutit o!'
Accorutting Bdrnares, and ASA 545
Auditing Fuir Value lvleasurcments and
Diulosures.

CHAPTER

Measurement

theory

159