Professional Documents
Culture Documents
Pertemuan 7-2007-Hitz-The Decision Usefulness of FairValue Accounting - A Theoretical
Pertemuan 7-2007-Hitz-The Decision Usefulness of FairValue Accounting - A Theoretical
To cite this article: Joerg-Markus Hitz (2007): The Decision Usefulness of Fair Value
Accounting – A Theoretical Perspective, European Accounting Review, 16:2, 323-362
This article may be used for research, teaching, and private study purposes.
Any substantial or systematic reproduction, redistribution, reselling, loan, sub-
licensing, systematic supply, or distribution in any form to anyone is expressly
forbidden.
The publisher does not give any warranty express or implied or make any
representation that the contents will be complete or accurate or up to
date. The accuracy of any instructions, formulae, and drug doses should be
independently verified with primary sources. The publisher shall not be liable
for any loss, actions, claims, proceedings, demand, or costs or damages
whatsoever or howsoever caused arising directly or indirectly in connection
with or arising out of the use of this material.
European Accounting Review
Vol. 16, No. 2, 323 –362, 2007
JOERG-MARKUS HITZ
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
ABSTRACT Regulators such as the SEC and standard setting bodies such as the FASB and
the IASB argue the case for the conceptual desirability of fair value measurement, notably on
the relevance dimension. Recent standards on financial instruments and certain non-financial
items adopt the new measurement paradigm. This paper takes issue with the notion of
decision usefulness of a fair-value-based reporting system from a theoretical perspective.
Emphasis is put on the evaluation of the theoretical soundness of the arguments put
forward by regulators and standard setting bodies. The analysis is conducted as economic
(a priori) analysis. Two approaches to decision usefulness are adopted, the measurement
or valuation perspective and the information perspective. Findings indicate that the
decision relevance of fair value measurement can be justified from both perspectives, yet
the conceptual case is not strong. The information aggregation notion that underlies
standard setters’ endorsement of fair value measurement turns out to be theoretically
restricted in its validity and applicability. Also, comparative analysis of fair value
accounting vs. historical cost accounting yields mixed results. One immediate implication
of the research – a condition for the further implementation of fair value accounting – is
the need to clarify standard setters’ notion of accounting income, its presumed
contribution to decision relevance and its disaggregation.
1. Introduction
This paper is motivated by the ongoing shift of financial reporting standards for
listed companies towards fair-value-based reporting, notably the increasing
Correspondence Address: Joerg-Markus Hitz, University of Cologne, Seminar fuer ABWL und fuer
Wirtschaftspruefung, Albertus-Magnus-Platz, 50923 Köln, Germany. E-mail: hitz@wiso.uni-koeln.de
amounts, that is, asset write-ups are not value-relevant. Summarizing the extant
empirical literature, the relevance of fair value measurement can only be sup-
ported for securities traded on highly liquid markets, while the evidence
reinforces the significance of the reliability objection both for financial and
non-financial assets.
Theoretical research so far has been relatively silent on the properties and
desirability of fair value measurement. While the informational quality of
market values is unassailable under conditions of complete and perfect
markets, the contribution of fair value measurement to valuation or contracting
purposes is unclear in a realistic setting (Beaver, 1998). Notably, the existing
body of research does not take issue with the theoretical assumptions and hypo-
theses underlying the fair value paradigm as articulated by standard setters.
This paper contributes to the literature on fair value accounting in two ways.
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
the price that would be received to sell an asset or paid to transfer a liability in an
orderly transaction between market participants at the measurement date.’ The
IASB framework at present has no definition of fair value, yet a uniform defi-
nition can be found on the standards level: ‘Fair value is the amount for which
an asset could be exchanged, or a liability settled, between knowledgeable,
willing parties in an arm’s length transaction.’1 In a current convergence
project, the IASB develops an International Financial Reporting Standard
(IFRS) on fair value measurements, which is based on SFAS 157.2
Taking into account the relevant interpretations, the FASB/IASB concept of
fair value is that of a specific hypothetical market price under idealized con-
ditions. More precisely, fair value is the exit market price that would result
under close-to-ideal market conditions, in a transaction between knowledgeable,
independent and economically rational parties, which interact on the basis of an
identical information set (complete information). The sharp distinction of fair
value and value in use clarifies that fair value measurement is not to include
entity-specific competitive advantages, that is, no private skills and no private
information (SFAS 157, para. C32; SFAC No. 7, para. 24 a; JWG, 2000, para.
4.5; IASB, 2006, paras. 42– 45).
The estimation of fair value follows, in principle, a three-tier hierarchy. The
governing principle is primacy of market-based measures – the refutable
notion that market prices or market data are more informative and reliable than
internal estimates. Thus, market prices represent the best estimate of fair value,
if market conditions satisfy the fair value definition. The relevant ‘quality’ of
market prices is assessed on the basis of the active market criterion, that is,
regular trading of the item on a sufficiently liquid market is required for the
market price to qualify as an estimate of fair value.3 If market prices do not
exhibit sufficient quality or are not available, the second level of the estimation
hierarchy requires to consider (modified) market prices of comparable items,
where comparability naturally refers to the cash flow profile. Only when such
prices cannot be used either, marking-to-market fails and fair value is mandated
to be estimated using internal estimates and calculations. This marking-to-model,
The Decision Usefulness of Fair Value Accounting 327
expectations of investors in the market concerning the cash flow pattern of the
asset or liability:
SFAC No. 7 appears to generalize this notion for any fair value, that is, model-
based estimation is assumed to arrive at hypothetical prices with similar informa-
tional properties.7 It is hypothesized that investors can extract these implicit con-
sensus expectations from market prices in order to revise and improve their own
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
vague notion of ‘nondistortion’ (Bevis, 1965, p. 104), which had increasingly been
perceived as a pretext for discretionary definitions of balance sheet positions.9
Additionally, researchers and regulators felt uncomfortable with a balance sheet
that had no informative purpose of its own.10 The FASB therefore adopted the
new asset-liability approach with SFAC No. 3 in 1980 (SFAC No. 6 in 1985),
which implements economics-based definitions of assets and liabilities that
refer to future economic benefits and outflows of those benefits, respectively,
and links income strictly to changes in net assets.11
Over the years, there has been considerable debate on whether the asset-liab-
ility approach requires measurement based on current values rather than histori-
cal cost, which resulted in controversies notably at the FASB (Miller, 1990,
p. 28). In any event, the approach stresses the role of the balance sheet as a
source of decision useful information and therefore provides a conceptual under-
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
pinning for the initiation of the fair value paradigm, which was stimulated by the
specific problems involved in accounting for financial instruments.
A critical event triggering the shift towards the fair value paradigm was the
Savings-and-Loans (S&L) Crisis in the USA during the 1980s, which laid
open the deficiencies of the prevalent reporting system based on the historical
cost/matching paradigm. It resulted in regulatory action by the SEC, which
among other things advised the FASB to develop a standard on accounting for
certain debt securities at their market value instead of amortized cost (Wyatt,
1991; Cole, 1992; White, 2003). The underlying notion was that historical cost
accounting had hindered proper identification of the financial status of S&L;
notorious practices were the designation of securities as investments in order
to avoid write-offs, accompanied by the realization of gains on securities
trading above their book values (‘cherry picking’ or ‘gains trading’). Despite
its limited scope, this initiative represented a major evolution in accounting
thought on the regulatory level.12
The immediate regulatory reaction in the wake of the S&L crisis represents the
starting point for the implementation of fair value measurement and the evolution
of the fair value paradigm both in FASB and IASB standards. Starting out as a
special regulation for certain securities, fair value measurement was soon ident-
ified as the most relevant attribute for financial instruments. Full fair value
accounting for financial instruments was advocated by the IASC in its 1997 dis-
cussion paper, which represented a basis for the Joint Working Group Draft Stan-
dard in 2000. Paralleling this process was the tentative introduction of fair value
for non-financial items, where SFAC No. 7 on the present value measurement of
fair value constituted a landmark conceptual step. Notably, SFAC No. 7 appears
to generalize the fair value paradigm for any market value satisfying the fair
value definition – be it market-order model-based – and arrives at a fundamental
conclusion:13
With the adoption of SFAC No. 7, the FASB lends constitutional character to the
fair value measurement objective: given the normative function of the conceptual
framework, fair value measurement is an alternative to be considered in any
future standard setting initiative. The latest step in the evolution of the fair
value paradigm has been taken only recently. SFAS 157, Fair Value Measure-
ments, gives ample guidance on the definition and estimation of fair value,
thus providing an authoritative (Level-A-GAAP) principles-based concept
which applies to any standard employing the fair value measurement objective.
In its current project on fair value measurements, the IASB adopts the SFAS
157 fair value hierarchy.
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
preclude the recognition of fair value gains beyond the cost ceiling. Specifically,
fair value represents the relevant impairment measure for goodwill acquired in a
business combination, certain intangible assets (SFAS 142) and long-lived assets
(SFAS 144). IAS 36 requires similar impairment rules, with recoverable amount,
the higher of value in use and fair value less cost to sell, as the relevant measure-
ment attribute. IFRS standards, on the other hand, go far beyond FASB
provisions. The revaluation model, which can be optionally applied in accounting
for property, plant and equipment (IAS 16) and for actively traded intangibles
(IAS 38), requires full fair value measurement, with remeasurement gains
beyond historical cost taken to revaluation surplus (other comprehensive
income). The fair value model provided optionally for investment property
(IAS 40) and compulsory for biological assets (IAS 41) requires full fair value
accounting with gains and losses taken directly to income.
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
. Does fair value represent decision useful information? Is there a valid theoreti-
cal background to standard setters’ promotion of fair value measurement?
. Should fair values be merely disclosed, or is there a conceptual case for recog-
nition in basic financial statements?
. Do revaluation gains from fair valuation represent components of income or
should they be recognized outside earnings?
. What are the basic properties of fair value income and its contribution to the
decision usefulness objective?
Although the use of current values has been a long-standing issue both in
accounting research and regulation,14 with many valuable insights produced
notably by the theoretical ‘a priori research’ debate,15 this strand of literature
shall not be pursued here for two reasons that constitute the uniqueness of the
fair value debate. Firstly, the previous sections have shown that fair value is
one specific current value concept which has only recently entered the accounting
stage, with a unique definition and set of estimation rules. Secondly, in contrast to
concepts previously discussed in the literature, for example, current value,
deprival value or net realizable value, fair value measurement is grounded in a
specific theoretical reasoning embodied by the paradigmatic information aggre-
gation hypothesis, which rests on the decision usefulness concept. This paper
therefore takes issue with the decision usefulness of reporting fair value
measures, with emphasis given to the validity of its paradigmatic foundations.
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
conceptual evaluation. For the purposes of this paper, two concepts of decision
usefulness from an information perspective shall be distinguished:
of our analysis lend themselves to inferences for accounting regulation and there-
fore also contribute to the standard setting literature and the related debate on fair
value accounting. The results of a priori economic analysis on the contribution of
alternative regulations to conceivable measures of decision usefulness potentially
provide supplemental, new information to standard setting bodies. As such, they
improve standard setters’ knowledge and are therefore capable of improving stan-
dard setting decisions. Given this ‘decision-model approach’ to the relationship
between accounting research and regulation (Beaver and Demski, 1974, pp.
175 –177), one task of researchers is to inform on theoretical results and their
implications, while ultimate evaluations and (value) judgments are solely left
to standard setting bodies.
For the purpose of drawing standard setting inferences, the correspondence of
the theoretical perspectives employed in this analysis and FASB’s/IASB’s
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
firm’s assets and liabilities. It allows for the evaluation of the informational prop-
erties of fair value from a measurement and from an information perspective.
Plus, it is consistent with the view inherent in the fair value paradigm: the funda-
mental information aggregation hypothesis regards fair value per se rather than
questions of (aggregated) fair value accounting or even fair value income. This
part of the analysis does not extensively address the comparative suitability of
fair value vis-à-vis historical cost. Since the conceptual basis for the relevance
of historical cost rests on the revenue –expense approach and the corresponding
income concept, there is no similar claim to the relevance of (aggregated) histori-
cal cost disclosures. The focus of the fair value paradigm on the relevance dimen-
sion, which is reflected in the choice of the evaluative criteria employed here, also
implies that aspects of reliability are only addressed as they come up, rather than
in a systematic fashion. Also, cost – benefit considerations are outside the scope of
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
the analysis.
Comparative aspects are addressed in the second part of the analysis (Section 5),
where accounting measurement is considered and the conceptual merits of
balance sheet and income concept under fair value accounting and under histori-
cal cost accounting are evaluated. Again, measurement and information perspec-
tive provide the framework for evaluation of decision usefulness. As abstract
notions, these perspectives will be further refined here in order to arrive at eva-
luative criteria for the comparative analysis of fair value income vs. historical
cost income. Again, aspects of reliability are not addressed.
X
N X
N
Vt ¼ VIUnt ¼ FVnt :
n¼1 n¼1
X
N X
N
Vt ¼ VIUnt ¼ FVnt þ gt :
n¼1 n¼1
To summarize, the conceptual case for fair value measurement from a rigid
measurement perspective can only be made for an idealized scenario of complete
and perfect markets which has no demand for financial reporting. For a real-world
setting, even if well-developed markets are assumed, fair value measurement
338 J.-M. Hitz
combine the (M ) fair values of these activities and the present value of cash
flows from remaining activities (c) (discounted at the cost of capital k) in order
to arrive at firm value:
X
M X
T
Et ½ct
Vt ¼ FVnt þ :
n¼1 t¼tþ1
(1 þ k)tt
Ignoring costs, this separation model illustrates how fair value information
improves decision making and is thus decision useful, if it allows for more
precise valuation. The approach is well established in financial analysis (e.g.
Penman, 2003, p. 455) and lends vindication, for example, to fair value account-
ing for investment property and, if one is inclined to assume separability, for
financial activities (Feltham and Ohlson, 1995). Yet, it not only requires separ-
ability and zero rents for the activities in question, but also high information
quality of fair value: fair value will only substitute subjective projections if inves-
tors accept it as a more accurate measure of the present value of future cash flows.
Also, a further condition is that cash flow projections can be performed for the
remaining activities with no loss in accuracy. Therefore, the descriptiveness of
the information aggregation assumption underlying the fair value paradigm is
vital. This will be further explored once the information perspective has been
considered.
gate the private information dispersed in the marketplace (v. Hayek, 1945,
p. 526) needs further examination. The informational quality of fair value as
market price and thus the validity of the paradigmatic information aggregation
assumption rests on the extent to which investors’ private information is factored
into the market price, that is, on the degree of market information efficiency in the
strong Fama sense. The modern theory of asset pricing under asymmetric infor-
mation, notably rational expectations equilibria and strategic trader models, gives
valuable insights into this question.
Application of the theory of rational expectations to asset pricing emphasizes
the dual role of prices: not only does the price system in equilibrium balance
supply and demand and clear the market; it also represents a source of infor-
mation for market participants, who extract from market prices knowledge
about other investors’ private information. Thus, a major result of this strand
of research is the information content of market prices: investors with individual
expectations (private information sets) will potentially exhibit different invest-
ment behaviour if they also observe market prices – the (costless) availability
of market price information is capable of inducing revisions in investment
decisions (Grossman, 1981, pp. 549 –554). A second important result concerns
the degree of informational efficiency, that is, the information set that can be
inferred from market prices. The Grossman paradox illustrates that perfectly
informative, ‘fully revealing’ prices cannot exist in an equilibrium with costly
information acquisition, because perfect inference from prices eliminates incen-
tives for private information collection, which in turn reduces the informative-
ness of prices (Grossman, 1976). The implication is that only where additional
noise inhibits the quality of prices as sufficient statistics for consensus expec-
tations will incentives for information acquisition prevail. This leads to the
paradox result that noise in the price system is a condition for its informativeness
– market efficiency in the strong sense cannot be accomplished. Noisy rational
expectations equilibria recognize these precepts and show that, given stochastic
noise, the informational quality of market prices, that is, the degree of private
investor information diffusion and aggregation, increases with reduced investor
The Decision Usefulness of Fair Value Accounting 341
risk aversion and with the precision of their private information, whereas it is
reduced with the cost of private information acquisition and with noise (Grossman
and Stiglitz, 1980; Hellwig, 1980; Diamond and Verrecchia, 1981; Verrecchia,
1982).
A different theoretical branch, the so-called strategic trader models, yields
additional results on the determinants of market prices’ information quality.
Here, the focus lies on the strategic implications of the use of private information
by insiders, especially on the factors which determine the speed and amount at
which such insiders give their information into the market and allow them to
be factored into the price system. In a seminal model, Kyle (1985) shows that
in Bayes – Nash equilibrium, the ‘aggressiveness’ of the insider’s use of his
private information critically depends on the amount of non-information-based
trading, which provides noise and thus camouflage for the insider. Market
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
makers, on the other hand, anticipate the insider’s strategy and make price adjust-
ments that are inversely correlated with the amount of noise trading, that is, the
possibility to compensate losses from trading with the insider with gains from
trading with uninformed market participants. This result implies that the
market liquidity in its price dimension (the price reaction to an order) evolves
endogenously as a reaction to insider trading and is thus a theoretically sound
indicator of adverse selection or information quality of the price system
(Glosten and Milgrom, 1985; Kyle, 1985).20 Follow-up models refine these
results and demonstrate that the quality of market prices increases with the com-
petition among insiders and the precision of their private signals, and decreases
with their risk aversion and with the volume of noise trading (Kyle, 1984;
Subrahmanyam, 1991; Holden and Subrahmanyam, 1992; Vives, 1995).
In conclusion, the theory of asset pricing under asymmetric information yields
several insights into the informational properties of fair value estimated as market
price. At the outset, the Grossman paradox shows that this fair value cannot be
fully informative: unbiased consensus expectations cannot be inferred. Theoreti-
cal models illustrate the factors that determine the quality of partially revealing
market prices. More recent insights from behavioural finance theory suggest
that in addition to noise, irrational market behaviour is a factor reducing the infor-
mational quality of market prices (Shleifer, 2000). Thus, the paradigmatic infor-
mation aggregation assumption holds roughly only for specific assets traded on
organized, highly liquid markets. In contrast, markets for positions not traded
on organized exchanges can be characterized as search markets (Krainer and
LeRoy, 2002). Under such circumstances, market prices normally cannot be
interpreted in the paradigmatic sense, since they result from specific transactions
between two parties and rather indicate value in use than aggregate the consensus
expectations of numerous market participants.
In addition, rational expectations equilibrium models spell out and confirm the
‘learning from prices’ assumption, according to which investors infer infor-
mation about the probability distribution of cash flows. These models thus
provide a backing for the notion of inferring decision useful information from
342 J.-M. Hitz
4.3.2. Marking-to-model
With the move from market price valuation to the modelling of a synthetic market
value, fair value becomes a hypothetical market price under ideal rather than
idealized conditions. This is due to the neoclassical basis, the strict assumptions
underlying contemporary pricing models. The Capital Asset Pricing Model
(CAPM) is representative of these models and illustrates the ideal character of
resulting estimates. It constitutes the foundation of present value calculations
and is explicitly suggested by FASB and IASB as valuation method (SFAC
No. 7, paras. 62 –71; IAS 36, para. A17). One fundamental assumption
The Decision Usefulness of Fair Value Accounting 343
not only generally, but also specifically for fair value measurement, supports the
relevance of the reporting form (Beatty et al., 1996; Ahmed et al., 2004).
It has been pointed out that the paradigmatic foundations of fair value measure-
ment refer to fair value per se and therefore do not support any specific form of
aggregation or presentation. Notably, no arguments are given for fair value
accounting, since the disclosure of fair values would suffice to benefit from the
alleged informational properties. Yet, fair value is increasingly being
implemented for balance sheet and income measurement. Assuming a positive
role for historical-cost-based financial statements, notably for contracting pur-
poses (Watts, 2003), implementation of fair value accounting therefore requires
theoretical support beyond the informational quality of fair value per se. It is the
aim of the following sections to analyse the properties and potential decision use-
fulness of fair value accounting in a comparative setting. For this purpose, the
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
Yet, for conceptual reasons, it cannot eliminate this gap, which consists of two
further components: the fair values of identifiable, yet not recognizable positions,
for example, certain internally generated assets (‘recognition gap’), and the fair
values of the remaining, non-identifiable factors reflected in internally generated
goodwill (‘goodwill gap’). Obviously, fair value accounting is neither conceived
for nor conceptually capable of measuring directly the value of the firm.
Again, application of the less restrictive decision-model approach leads to a
more positive evaluation. By assuming clean surplus accounting and rewriting
the investor’s cash flow valuation calculus as residual income model, the
notion of book value of equity as heuristic measure of a fraction of enterprise
‘value’ can be supported. The residual income valuation formula views firm
value as the sum of current book value of equity plus the present value of
future residual (abnormal) earnings (Edwards and Bell, 1964; Ohlson, 1995).
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
of fair value income are not reflected in the paradigmatic foundations of fair value
measurement, which rest solely on a stocks perspective. Reception in the
accounting research literature has been equally sparse, since most empirical
studies focus the value relevance of fair value disclosures, and analytical
papers so far have rarely taken issue with fair value income. These observations
contrast starkly with the prominent role of earnings in the capital market, which is
normally reflected and recognized by the theoretical literature (Nichols and
Wahlen, 2004). Additionally, the most contentious aspects of the debate on
fair value accounting centre on its implications for earnings: while the fundamen-
tal questions refer to recognition vs. disclosure, the qualification of revaluation
gains and losses and the disaggregation of income (performance reporting), the
practical, sometimes even political, debate revolves around earnings volatility.
Opponents of fair value income maintain that current valuation leads to increased
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
relevant outside of the steady state, that is, for firms with increasing or declining
net assets (Zhang, 2000, p. 132).
To conclude, fair value income represents no unbiased indicator of economic
income, since, given growth in operations, both earnings measures differ system-
atically, where the differences increase with the gap between firm value and the
book value of equity. Although fair value earnings are conceptually closer to
economic income than historical cost earnings, the systematic difference prohi-
bits support from a strict measurement perspective. Apparently, proponents of
fair value income who insist on its capability to depict ‘economic reality’ (e.g.
Sprouse, 1987, p. 88) implicitly take on such an economic perspective on
income measurement. However, the systematic differences indicate that this
argument is ill-founded, because mismatching occurs due to unrecognized
assets and goodwill and impairs fair value income’s capacity to accurately
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
Thus, for assets whose cash flow patterns do not satisfy this property – this will
be the case for any asset subject to wear and tear – the expected value follows a
trend, with deviations from this trend occurring randomly. Accordingly, the
change in fair value consists of an expected and an unexpected component, the
fair value shock. The conjecture of random movements in fair value is apparently
too crude a qualification, since it applies only to a fraction of the revaluation
difference. This is further illustrated by looking at fair value income, which is
defined as the change in fair value plus the cash flow realization and equally con-
sists of an expected and an unexpected component:
xFV
t ¼ FVt FVt1 þ ct
¼ k|fflfflfflfflfflffl{zfflfflfflfflfflffl}
FVt1 þ ½ct Et1 (ct ) þ ½FVt Et1 (FVt ) :
|fflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflffl} |fflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflffl}
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
expected fair value income cash flow shock fair value shock
The unexpected cash flow component is strictly transitory, because fair value is
independent of the current cash flow realization. Yet, the fair value shock is partly
persistent, since it directly bears on the current fair value, which is multiplied
with the market return to yield next period’s expected income: the unexpected
change in fair value (revaluation gain/loss) persists at a rate k into future
periods. Additionally, systematic deviations from expected fair value income
occur when the position is associated with rents. If the firm can put the position
to a superior use relative to the market or has private information concerning its
prospective cash flows, the ‘unexpected’ component of fair value income will be
serially correlated and thus have predictive ability.
To conclude, the Samuelson theorem only superficially presents a theoretic
backing for the conjecture of random moves in fair values and thus the distortion
of income’s predictive ability. Except for those positions which yield cash flows
equal to expected return, for example, shares and non-depletable assets, fair value
remeasurement gains and losses will be correlated in time, despite market effi-
ciency. A fair value income measure which incorporates the cash flow component
will exhibit persistence at the rate of the expected market return k, since fair value
shocks feed into current fair value and thus directly bear on expected return.
Persistence of fair value income. The previous results apply to single assets
only and are not of a comparative nature. Therefore, the next step is to compare
fair value income to the transaction-based concept. Since such earnings numbers
are the result of a complex measurement process and also include revenues and
gains deriving from transactions and events not recognized on the balance sheet,
a thorough analytical investigation is not undertaken. Instead, we look at a sty-
lized scenario to delineate some basic comparative results. For this purpose, a
single-asset firm is assumed which produces one good for a limited number of
periods (i.e. reinvestment decisions are not considered) and sells a uniform
amount every period, capitalizing on a competitive advantage that enables the
firm to demand a premium price.
352 J.-M. Hitz
In a scenario where all expectations are precisely met, fair value income will,
over time, exhibit a strictly decreasing pattern. Despite uniform cash flows, the
interest component of fair value, which offsets the cash flow component, will
decline, leading to an increase in ‘fair value depreciation’. Transaction-based
income, on the other hand, is uniform due to straight-line depreciation and
thus highly persistent, which suggests superior predictive ability.25 These,
however, are rather crude qualifications that only serve as a starting point. Predic-
tive ability matters where expectations are not met, that is, where economic
shocks occur. Therefore, the implications of a sales shock and an interest rate
shock are analysed.
A sales shock, the increase in goods produced and sold that results from an
unexpected rise in demand, causes an unexpected increase for both event-
based fair value income and transaction-based historical cost income. In so far,
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
both concepts reflect the ‘good news’. Since historical cost income returns to a
steady path in the following period, it suggests greater persistence. Notably, a
persistent sales shock will result in a fully persistent income shock, since the
additional revenue will reoccur over the following periods. Clearly, this property
of historical income is what many proponents of transaction-based income recog-
nition have in mind, because fair value income is less stable: it exhibits a one-
time spike in the period of the economic shock and then declines in the following
periods. If one subscribes to the strict definition of useful income as persistent
income, a case can be made for transaction-based accounting. Put differently,
the volatility criticism can be supported from such a perspective.
However, putting aside the earnings persistence criterion, a second glance at
the time-series behaviour of these two earnings concepts reveals a property of
fair value income so far hardly encountered in the literature. Since transaction-
based income will only recognize the effects of the economic shock on current
period’s sales, it cannot discriminate one-time shocks from lasting, that is, per-
sistent effects. This lack of responsiveness contrasts sharply with fair value
income, which will c.p. react the stronger, the more persistent the economic
shock. Since the fair value shock captures the revisions of cash flow expectations
for all future periods, it will correlate with the persistence of the economic (sales)
shock. The fair value shock as a component of fair value income, unlike unex-
pected historical cost income, discriminates the persistence of economic
shocks and thus allows for a more precise state partition. That is, despite its
lack of earnings persistence according to the conventional definition, fair value
income appears to represent the finer information system.
This property is further illustrated by considering a lasting interest rate shock,
the unexpected negative shift of the term structure of interest rates, which mod-
ifies cost of capital and therefore the discount factor underlying the fair value of
the firm’s asset. This represents an economic event that directly impacts on firm
value, yet leaves the cash flows unchanged. Historical cost income determination
therefore completely ignores the shift in interest rates, whereas fair value income
rises unexpectedly. Again, transaction-based income is more steady and
The Decision Usefulness of Fair Value Accounting 353
persistent, whereas the rise in fair value income is followed by declines due to the
interest rate effect, that is, it is negatively correlated. Yet, unlike historical cost,
fair value income signals the occurrence of a valuation-relevant event and thus
transports more information.
In conclusion, the fundamental conjectures against the predictive ability of fair
value income can be supported if one employs the concept of earnings persistence.
Fair value income incorporates economic shocks (1) more extensively, since their
implications for all future periods are immediately recognized, and (2) more com-
pletely, because cash-flow-irrelevant shocks are also reflected. This greater respon-
siveness to valuation-relevant events impairs the steadiness of the earnings stream,
that is, the autocorrelation of earnings (shocks). On the other hand, in informing
more precisely and more thoroughly about these relevant events, fair value
income represents the finer information system. Put differently, fair value income
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
(1) There is a theoretical case for the disclosure of prices taken from organized,
sufficiently liquid markets, since these allow for the inference of the aggre-
gated market consensus expectations concerning amounts, timing and uncer-
tainty of future cash flows. The relevance of fair value disclosures for traded
financial instruments thus receives support. Given its conceptual merits for
financial income determination and recognition of hedging activities, full
fair value accounting for financial instruments also appears to be the recom-
mendable path, despite reliability concerns for non-publicly traded instru-
ments and distortions vis-à-vis the economic income model.
(2) Since fair value measurements based on valuation models do not inform
about consensus expectations, the conceptual backing appears particularly
weak for fair valuation of non-financial items. In addition to this relevance
objection, empirical evidence supports the notion of grave reliability con-
cerns for fair values not taken from active markets. At present, no sound
The Decision Usefulness of Fair Value Accounting 355
IASB and FASB appear to share some of these implications. As pointed out,
there is a general consensus on the conceptual merits of a full fair value account-
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
ing model for financial instruments (implication (1)). However, the recent
curbing of the fair value option in IAS 39 illustrates the difficulty of implement-
ing this view, notably due to reliability concerns. As for implication (2), at least
the IASB with its more progressive use of fair value for non-financial items
appears not to consent. While the adoption of the measurement guidance in
SFAS 157 by the IASB in its current project may increase the reliability of fair
value measurements, the conceptual criticism of the information aggregation
hypothesis remains unchallenged. If there were other motives driving the use
of fair values for non-financial items, for example, a theory of private information
revelation by management, that is, a concept of value in use rather than hypothe-
tical market value, they should be stated explicitly, thus limiting the scope and
generality of the fair value paradigm. The same applies to the position of the
FASB, which is more reluctant to use fair value for non-financial assets, yet,
notably in SFAC No. 7, provides the conceptual basis for generalizing the infor-
mation aggregation hypothesis. In a similar vein, the requirement to develop and
communicate the concept of informative earnings pursued (implication (3)) has
hit the agenda in the wake of the discussion of financial reporting quality trig-
gered by the Enron failure, which among other things resulted in a commitment
to more ‘principles-based’ standard setting. IASB and FASB will address the
income concept in the course of their conceptual framework project, while the
joint project on Financial Statement Presentation (formerly Reporting Perform-
ance) is concerned with earnings disaggregation. For the latter, the observations
on fair value income made here indicate that there is a case for separately disclos-
ing revaluation gains and losses that result from revisions in cash flow expec-
tations (fair value shocks).
Acknowledgements
I am indebted to Christoph Kuhner for his critical and insightful comments at
various stages of this project. Also, the helpful suggestions of two anonymous
reviewers and the journal’s previous editor, Kari Lukka are gratefully
356 J.-M. Hitz
Notes
1
See, for example, IAS 39, para. 9; IAS 41, para. 8; IFRS 3, Appendix A; IFRS 4, Appendix
A. The broad correspondence of the two standard setters’ concepts is demonstrated by the
IASB’s reaffirmation to adopt SFAS 157 as a basis for the project on Fair Value Measurements.
2
In December 2006, the IASB issued a Discussion Paper which states broad agreement with
SFAS 157. Thus, final convergence concerning the definition and estimation of fair value is
on the horizon. However, since some IAS/IFRS standards appear to employ an entry value
notion of fair value, the IASB considers scope limitations for an IFRS on fair value measure-
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
ments, or, alternatively, introduction of an entry price measurement attribute for those stan-
dards. See IASB (2006, paras. 12–17). For an analysis of the current IAS/IFRS fair value
guidance, see Cairns (2006, pp. 7–10).
3
See the definition in SFAS 157, para. 24. Under IFRS, a uniform definition of an active market
is applied on the standards level; see IAS 36, para. 6; IAS 38, para. 8; IAS 41, para. 8.
4
Thomas S. Kuhn in his ‘Structure of Scientific Revolutions’ extensively discusses this term in
the context of scientific methodology and develops his influential theory of the process and
drivers of paradigm shifts. While Kuhn’s ideas have been subject to contentious debate, with
himself backtracking on the definition, the general idea of a paradigm as a set of shared
beliefs appears useful for tentatively conceptualizing the theoretical basis of fair value
measurement.
5
The implementation of the decision usefulness objective drew from the Trueblood Report.
However, the foundations had been developed much earlier both in the academic and the stan-
dard setting sphere. Significant contributions involve the AAA’s ASOBAT and APB Statement
No. 4. For an historical overview see Hendriksen and van Breda (1991, pp. 92– 115, 126–131).
6
See also SFAS 115, para. 40; SFAS 107, para. 39; SFAS 133, para. 220; and SFAS 157, para.
C32. For similar conjectures in IASB pronouncements, see, for example, IAS 36, para. BCZ11;
IAS 40, para. 40, B36; IASC (1997, ch. 5, paras. 2.6–2.12); and, more recently and broadly in
the context of initial recognition, IASB (2005, paras. 99, 101 –104).
7
See SFAC No. 7, especially the section on ‘Present Value and Fair Value’ (paras. 25 –38), and
the observations on the evolution and implementation of the fair value paradigm in the follow-
ing Section 2.3.
8
For an even earlier exposition of a similar approach, see Schmalenbach (1919) on the concept of
a ‘dynamic balance sheet’.
9
In an influential paper, Sprouse (1978) termed those balance sheet positions that are mere by-
products of the matching process and do not conform to notions of assets or liabilities ‘what-
you-may-call-its’ (p. 69). He observes: ‘Under the revenue/expense view, (. . .) what constitu-
tes “proper matching” and “nondistortion” is very much in the eyes of the beholder (. . .)
“Matching costs and revenues” is too often an attractive but empty slogan rather than a mean-
ingful concept that one can look to for guidance.’ See also Gellein (1992, p. 198) and Schuetze
(2001, pp. 9– 11).
10
The perception thus was that under a revenue–expense approach, the balance sheet merely
served as ‘mausoleum for the unwanted costs that the double-entry system throws up as regret-
table by-products’ (Baxter, 1977, p. x).
11
See Robinson (1991, p. 110). For the respective definitions, see SFAC No. 6, paras. 25– 43, 70,
78–89; IASB Framework, paras. 53–64, 70 (Nobes, 2001, p. 9). The validity and necessity of
the asset-liability approach as ‘conceptual anchor’ has only recently been reaffirmed by the SEC
(SEC, 2003, p. 10).
The Decision Usefulness of Fair Value Accounting 357
12
Former FASB member Arthur Wyatt refers to it as ‘possibly the most significant initiative in
accounting principles development in over 50 years’ (Wyatt, 1991, p. 80), a notion emphasized
by the testimony of SEC General Counsel James Doty to the US Senate, who made it clear that
‘the time has run out on “once-upon-a-time-accounting”‘.
13
A similar reasoning can be found in the IASB’s 2005 Discussion Paper on Measurement Bases,
which emphasizes that the relevance of fair value as a measurement attribute for initial recog-
nition stems from its ‘market value properties’, which are assumed to hold in principle for any
fair value, that is, prevail irrespective of its estimation. See IASB (2005, paras. 111, 227– 231).
14
For instance, market valuation of long-lived non-financial assets was well established in the
USA prior to the Great Crash and the ensuing security market regulation in the 1930s
(Walker, 1992, pp. 4–8). Although the SEC later prohibited such practice, the issue of
market value accounting for certain items was discussed regularly in the USA as well as in
other countries (Wyatt, 1991; Christie, 1992, p. 97).
15
For an overview see Nelson (1971). The debate can be traced back as far as to Canning (1929) or
Simon (1899). Later, it focused on the scope of current value measurement and the selection of
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
the appropriate measurement basis. Given the diversity of this literature, only some of its more
famous contributors are mentioned: for example, Chambers (1965) as an advocate of exit value,
Revsine (1970) and Sterling (1981) as advocates of entry value, Edwards and Bell (1964) and
their concept of business profit.
16
While Beattie (2002, p. 109), Demski et al. (2002, p. 163) and Scott (2003, p. 174) perceive the
move towards fair value measurement as a renaissance of the measurement perspective, Beaver
(1998, p. 4), for instance, frames this development in an information perspective context.
17
A thorough textbook description of the information economics approach to financial reporting is
given by Christensen and Demski (2003).
18
Thus, we do not subscribe to the implications of the impossibility result (Demski, 1973): the
denial of the usefulness of applying conceptual criteria on the grounds of the specifity of indi-
vidual decision contexts employs a Paretian notion of economic efficiency which is not ade-
quate for economic analysis. Rather, economic analysis requires an assessment of the welfare
implications of different alternatives which can only be properly evaluated on the basis of
Kaldor–Hicks efficiency. While Pareto efficiency prescribes that no user will suffer a reduction
in welfare from a new accounting regulation and at least one user will experience an increase in
personal welfare, Kaldor–Hicks efficiency is less restrictive in that it requires an increase in
aggregate welfare, that is, it implies that those users better off can compensate those users
that suffer decreases in welfare. See Posner (1998, p. 16); in the context of normative evaluation
of accounting alternatives Beaver and Demski (1974, p. 174), Cushing (1977, p. 311) and
Ohlson (1987, pp. 2, 6).
19
It is not the aim of this paper to give a thorough discussion on the merits of normative vs. posi-
tive research approaches. Based on the arguments given, it is simply assumed that there is a case
for a priori reasoning, especially with respect to standard setting questions. For a more general
discussion, see Cushing (1977), Christenson (1983), and Watts and Zimmerman (1990). For an
assessment of potential standard setting implications of empirical research, especially value-rel-
evance research, see Holthausen and Watts (2001).
20
Curiously, the active market criterion outlined in IFRS standards as a means for discriminating
sufficiently informative market prices (level one estimates, see note 2) exclusively refers to the
time dimension of liquidity, that is, the speed with which a transaction partner can be found,
rather than the price dimension as theoretically valid indicator of information quality, which
can be measured, for example, on the basis of bid–ask spreads.
21
The latter also applies to a traditional mixed model, where assets carried at historical cost are
mingled with assets valued at a lower current value. However, as pointed out earlier, a stand-
alone analysis of the informational properties of historical cost-based measures and thus,
balance sheet captions, appears misguided since, unlike fair values, there is no explicit claim
to approximating economic value and aggregating cash flow information.
358 J.-M. Hitz
22
The (re-)emergence of the EBO model can therefore be seen as a ‘renaissance’ of the measure-
ment perspective (Beattie, 2002, p. 109).
23
For the distinction between economic income in a narrow sense and this ‘economic profit’ see
Christensen and Demski (2003, pp. 40, 50).
24
Although, as pointed out, the frameworks elaborate no concept of informative earnings, several
paragraphs hint at the notion of persistent earnings. See, for example, SFAC 1, para. 44; SFAC
5, para. 31; IASB Framework, paras. 28, 72 –80.
25
These results are turned upside down for residual income. Fair value residual income is constant
in time and reflects the competitive advantage, that is, the fraction of sales that is earned on top
of market expectations. Transaction-based residual income, on the other hand, increases in time
and thus appears less indicative of the competitive advantage.
References
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
Ahmed, A. S., Kilic, E. and Lobo, G. (2004) Does geography matter? Evidence from relative value-
relevance of banks’ recognized versus disclosed derivative financial instruments, Working Paper.
Barker, R. (2004) Reporting financial performance, Accounting Horizons, 18(2), pp. 157 –172.
Barlev, B. and Haddad, J. R. (2003) Fair value accounting and the management of the firm, Critical
Perspectives on Accounting, 14(4), pp. 383 –415.
Barth, M. E. (2000) Valuation-based research: implications for financial reporting and opportunities
for future research, Accounting and Finance, 40, pp. 7–31.
Barth, M. E. and Clinch, G. (1998) Revalued financial, tangible, and intangible assets: associations
with share prices and non-market-based value estimates, Journal of Accounting Research,
36(Suppl.), pp. 199 –233.
Barth, M. E. and Landsman, W. R. (1995) Fundamental issues relating to using fair value accounting
for financial reporting, Accounting Horizons, 9(4), pp. 97–107.
Barth, M. E., Beaver, W. H. and Landsman, W. R. (1996) Value-relevance of banks’ fair value dis-
closures under SFAS No. 107, Accounting Review, 71(4), pp. 513–537.
Barth, M. E., Beaver, W. H. and Landsman, W. R. (2001) The relevance of the value relevance litera-
ture for financial accounting standard setting: another view, Journal of Accounting and Econ-
omics, 31, pp. 77–104.
Baxter, W. T. (1977) Introduction, in: W. T. Baxter and A. S. Davidson (Eds) Studies in Accounting,
3rd edn (London: Institute of Chartered Accountants in England and Wales).
Beattie, V. (2002) Traditions of research in financial accounting, in: B. Ryan, R. Scapens and
M. Theobald. Research Method and Methodology in Finance and Accounting, pp. 94–113
(London: Thomson Learning).
Beatty, A. (1995) The effects of fair value accounting on investment portfolio management: how fair
is it?, Federal Reserve Bank of St. Louis Review, January/February, pp. 25–39.
Beatty, A., Chamberlain, S. and Magliolo, J. (1996) An empirical analysis of the economic impli-
cations of fair value accounting for investment securities, Journal of Accounting and Economics,
22, pp. 43–77.
Beaver, W. H. (1998) Financial Reporting – An Accounting Revolution, 3rd edn (London: Prentice
Hall).
Beaver, W. H. (1999) Discussion of ‘On transitory earnings’, Review of Accounting Studies, 4, pp.
163–167.
Beaver, W. H. (2002) Perspectives on recent capital markets research, Accounting Review, 77(2), pp.
453–474.
Beaver, W. H. and Demski, J. S. (1974) The nature of financial accounting objectives: a summary and
synthesis, Journal of Accounting Research, 12(3), pp. 170 –187.
Beaver, W. H. and Demski, J. S. (1979) The nature of income measurement, Accounting Review,
54(1), pp. 38 –46.
The Decision Usefulness of Fair Value Accounting 359
Beaver, W. H. and Venkatachalam, M. (2000) Differential pricing of components of bank loan fair
values, Journal of Accounting, Auditing and Finance, 15(1), pp. 41–67.
Beaver, W. H., Kennelly, J. W. and Voss, W. M. (1968) Predictive ability as a criterion for the
evaluation of accounting data, Accounting Review, 43(4), pp. 675–683.
Benston, G., Bromwich, M., Litan, R. E. and Wagenhofer, A. (2003) Following the Money – The
Enron Failure and the State of Corporate Disclosure (Washington: Brookings Institution Press).
Bevis, H. W. (1965) Corporate Financial Reporting in a Competitive Economy (New York:
Macmillan).
Bromwich, M. (1977) The use of present value valuation models in published accounting reports,
Accounting Review, 52(3), pp. 587 –596.
Bromwich, M. (2004) Aspects of the future in accounting: the use of market prices and ‘fair values’ in
financial reports, in: C. Leuz, D. Pfaff and A. Hopwood (Eds) The Economics and Politics of
Accounting – International Perspectives on Research Trends. Gedenkschrift für Dieter
Ordelheide, pp. 32–57 (Oxford and New York: Oxford University Press).
Cairns, D. (2006) The use of fair value in IFRS, Accounting in Europe, 3, pp. 5–22.
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
Grossman, S. J. and Stiglitz, J. E. (1977) On value maximization and alternative objectives of the firm,
Journal of Finance, 32(2), pp. 389 –402.
Grossman, S. J. and Stiglitz, J. E. (1980) On the impossibility of informationally efficient markets,
American Economic Review, 70(2), pp. 393–408.
Hellwig, M. F. (1980) On the aggregation of information in competitive markets, Journal of Economic
Theory, 22, pp. 477 –498.
Hendriksen, E. S. and van Breda, M. F. (1991) Accounting Theory, 5th edn (Boston: Irwin McGraw-
Hill).
Hicks, J. R. (1946) Value and Capital (Oxford: Oxford University Press).
Holden, C. W. and Subrahmanyam, A. (1992) Long-lived private information and imperfect compe-
tition, Journal of Finance, 47(1), pp. 247–270.
Holthausen, R. W. and Watts, R. L. (2001) The relevance of the value-relevance literature for financial
accounting standard setting, Journal of Accounting and Economics, 31, pp. 3–75.
IASB (2005) Measurement bases for financial accounting – measurement on initial recognition, Dis-
cussion Paper prepared by the staff of the Canadian Accounting Standards Board, London.
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
IASB (2006) Fair value measurements. Part 1: Invitation to comment and relevant IFRS guidance,
Discussion Paper, London.
IASC (1997) Accounting for financial assets and financial liabilities, Discussion Paper issued for
comment by the Steering Committee on Financial Instruments, London.
Joint Working Group of Standard Setters (JWG) (2000) Draft Standard and Basis for Conclusions:
Financial Instruments and Similar Items (London).
Krainer, U. and LeRoy, S. (2002) Equilibrium valuation of illiquid assets, Economic Theory, 19,
pp. 223 –242.
Kyle, A. S. (1984) Market structure, information, futures markets, and price formation, in: G. Storey,
A. Schmitz and A. H. Harris (Eds) International Agricultural Trade, pp. 45– 64 (Boulder, CO
and London: Westview Press).
Kyle, A. S. (1985) Continuous auctions and insider trading, Econometrica, 53(6), pp. 1315–1335.
LeRoy, S. F. (1989) Efficient capital markets and martingales, Journal of Economic Literature, 27, pp.
1583– 1621.
Liang, P. J. (2001) Recognition: an information content perspective, Accounting Horizons, 15(3), pp.
223–242.
Lipe, R. C. (1990) The relation between stock returns and accounting earnings given alternative infor-
mation, Accounting Review, 65(1), pp. 49–71.
Mauriello, J. and Erickson, J. (1995) Valuation risk and financial reporting, in: W. H. Beaver and
G. Parker (Eds) Risk Management – Problem & Solutions, pp. 171 –195 (New York: Prentice
Hall).
Miller, P. W. (1990) The conceptual framework as reformation and counterreformation, Accounting
Horizons, 4(4), pp. 23–32.
Nelson, K. (1971) A priori research in accounting, in: N. Dopuch and L. Revsine (Eds) Accounting
Research 1960–1970: A Critical Evaluation, pp. 3–19 (Urband, IL: University of Illinois).
Nelson, K. K. (1996) Fair value accounting for commercial banks: an empirical analysis of SFAS No.
107, Accounting Review, 71(2), pp. 161–182.
Nichols, D. C. and Wahlen, J. M. (2004) How do earnings numbers relate to stock returns? A review of
classic accounting research with updated evidence, Accounting Horizons, 18(4), pp. 263–286.
Nobes, C. (2001) Asset Measurement Bases in UK and IASC Standards (London: Certified Accoun-
tants Educational Trust).
O’Hara, M. (2003) Presidential address: Liquidity and price discovery, Journal of Finance, 58(4), pp.
1335– 1354.
Ohlson, J. A. (1987) On the nature of income measurement: the basic results, Contemporary Account-
ing Research, 4(1), pp. 1–15.
Ohlson, J. A. (1995) Earnings, book values, and dividends in equity valuation, Contemporary
Accounting Research, 11(2), pp. 661–687.
The Decision Usefulness of Fair Value Accounting 361
Ohlson, J. A. (1999) On transitory earnings, Review of Accounting Studies, 4, pp. 145 –162.
Park, M. S., Park, T. and Ro, B. T. (1999) Fair value disclosures for investment securities and bank
equity: evidence from SFAS 115, Journal of Accounting, Auditing and Finance, 14(3),
pp. 347–370.
Paton, W. A. and Littleton, A. C. (1940) An Introduction to Corporate Accounting Standards (Ann
Arbor).
Peasnell, K. V. (1977) The present value concept in financial reporting, Journal of Business Finance
and Accounting, 4(2), pp. 153– 168.
Penman, S. H. (1998) A synthesis of equity valuation techniques and the terminal value calculation for
the dividend discount model, Review of Accounting Studies, 2, pp. 303–323.
Penman, S. H. (2003) Financial Statement Analysis and Security Valuation, 2nd edn (New York:
McGraw-Hill).
Penman, S. H. and Sougiannis, T. (1998) A comparison of dividend, cash flow, and earnings
approaches to equity valuation, Contemporary Accounting Research, 15(3), pp. 343 –383.
Petroni, K. R. and Wahlen, J. M. (1995) Fair values of equity and debt securities and share prices of
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
property-liability insurers, Journal of Risk and Insurance, 62(4), pp. 719 –737.
Poon, W. W. (2004) Using fair value accounting for financial instruments, American Business Review,
22(1), pp. 39 –41.
Posner, R. A. (1998) Economic Analysis of Law, 5th edn (New York: Aspen Law & Business).
Ramakrishnan, R. T. S. and Thomas, J. K. (1998) Valuation of permanent, transitory, and price-irre-
levant components of reported earnings, Journal of Accounting, Auditing and Finance, 13(3), pp.
301– 349.
Razaee, Z. and Lee, J. T. (1995) Market value accounting standards in the United States and their sig-
nificance for the global banking industry, International Journal of Accounting, 30, pp. 208–221.
Revsine, L. (1970) On the correspondence between replacement cost income and economic income,
Accounting Review, 45(3), pp. 513 –523.
Robinson, L. E. (1991) The time has come to report comprehensive income, Accounting Horizons,
5(2), pp. 107 –112.
Ronen, J. (1974) Discounted cash flow accounting, in: J. J. Cramer, Jr. and G. H. Sorter (Eds) Objec-
tives of Financial Statements, pp. 143 –160 (New York: American Institute of Certified Public
Accountants).
Rubinstein, M. (1975) Securities market efficiency in an Arrow–Debreu economy, American Econ-
omic Review, 65(5), pp. 812 –824.
Samuelson, P. A. (1965) Proof that properly anticipated prices fluctuate randomly, Industrial Manage-
ment Review, 6, pp. 41–49.
Samuelson, P. A. (1973) Proof that properly discounted present values of assets vibrate randomly, Bell
Journal of Economics and Management Science, 4(2), pp. 369–374.
Schipper, K. and Vincent, L. (2003) Earnings quality, Accounting Horizons, Suppl., pp. 97 –110.
Schmalenbach, E. (1919) Grundlagen dynamischer Bilanzlehre, Zeitschrift fuer handelswissenschaft-
liche Forschung, 13, pp. 1– 50, 65–101.
Schuetze, W. P. (2001) What are assets and liabilities? Where is true north? (accounting that my sister
would understand), Abacus, 37(1), pp. 1–25.
Scott, W. R. (2003) Financial Accounting Theory, 3rd edn (Toronto: Prentice Hall).
SEC (2003) Study pursuant to Section 108(d) of the Sarbanes–Oxley Act of 2002 on the Adoption by
the United States Financial Reporting System of a Principles-Based Accounting System,
Washington.
Shleifer, A. (2000) Inefficient Markets: An Introduction to Behavioural Finance (New York: Oxford
University Press).
Simko, P. J. (1999) Financial instruments fair values and nonfinancial firms, Journal of Accounting,
Auditing and Finance, 14(3), pp. 247–274.
Simon, H. V. (1899) Die Bilanzen der Aktiengesellschaften und der Kommanditgesellschaften auf
Aktien (Berlin).
362 J.-M. Hitz
Sprouse, R. T. (1978) The importance of earnings in the conceptual framework, Journal of Accoun-
tancy, January, pp. 64–71.
Sprouse, R. T. (1987) Commentary on financial reporting, Accounting Horizons, 1(1), pp. 87– 90.
Sterling, R. R. (1981) Costs (historical versus current) versus exit values, Abacus, 17(2), pp. 93–129.
Subrahmanyam, A. (1991) Risk aversion, market liquidity, and price efficiency, Review of Financial
Studies, 4(3), pp. 417 –441.
v. Hayek, F. A. (1945) The use of knowledge in society, American Economic Review, 35(4), pp.
519–530.
Venkatachalam, M. (1996) Value-relevance of banks’ derivatives disclosures, Journal of Accounting
and Economics, 22, pp. 327–355.
Verrecchia, R. E. (1979) A proof of the existence of ‘consensus beliefs’, Journal of Finance, 34(4), pp.
957–963.
Verrecchia, R. E. (1982) Information acquisition in a noisy rational expectations economy, Econome-
trica, 50(6), pp. 1415–1430.
Vickrey, D. W. (1994) Measurement/valuation-related interpretations of fundamental accounting
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013