Pertemuan 7-2007-Hitz-The Decision Usefulness of FairValue Accounting - A Theoretical

This article was downloaded by: [University of Stellenbosch]
On: 01 May 2013, At: 15:41

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,
UK
European Accounting Review

Publication details, including instructions for authors
and subscription information:
http://www.tandfonline.com/loi/rear20
The Decision Usefulness of Fair

Value Accounting – A Theoretical
Perspective
a
Joerg-Markus Hitz
a
University of Cologne, Germany
Published online: 29 Jun 2007.
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
To link to this article: http://dx.doi.org/10.1080/09638180701390974
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-

and-conditions
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
The Decision Usefulness of Fair

Value Accounting – A Theoretical
Perspective
JOERG-MARKUS HITZ
Downloaded by [University of Stellenbosch] at 15:42 01 May 2013
University of Cologne, Germany
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
0963-8180 Print/1468-4497 Online/07/020323–40 # 2007 European Accounting Association

DOI: 10.1080/09638180701390974
Published by Routledge Journals, Taylor & Francis Ltd on behalf of the EAA.
324 J.-M. Hitz
importance of fair value as an accounting measurement attribute. Since the mid-

1980s, the US Financial Accounting Standards Board (FASB) and the Inter-
national Accounting Standards Board (IASB) have systematically substituted
market-based measures for cost-based measures. Starting out as a specific
remedy for the inequities of the reporting model for certain financial instruments,
fair value has manifested itself as the dominant measurement paradigm for finan-
cial instruments and, more recently, has increasingly been implemented for
measurement of non-financial items, for example, investment property under
IAS 40. The cost- and transaction-based reporting model is in decline, a new
market-value and event-based model on the rise, with dramatic implications
for the role and properties of balance sheet measurement and accounting income.
This shift in measurement paradigms is driven by the presumed decision rel-
evance of market-based measures. Both FASB and IASB stress the capacity of
market values to incorporate, in an efficient and virtually unbiased manner,

market consensus expectations about future cash flows. Opponents of fair
value measurement, on the other hand, criticize the questionable reliability of
fair value measures, especially for model-based estimates relying on manage-
ment’s expectations and projections. In particular, the implementation of fair
value as a balance sheet measure is the subject of intense discussion and
debate. The controversy about fair value accounting for financial instruments,
as recently highlighted by the rejection of IAS 39 (revised 2003) for full EU
endorsement, illustrates both conceptual and technical issues involved, for
example, the alleged distortion of earnings and aspects surrounding the
implementation of fair value hedging. Apparently, the debate is far from
resolved.
Prior empirical research on fair value measurement is mostly limited to finan-
cial instruments. Results so far support the incremental value relevance of fair
value disclosures for securities (Petroni and Wahlen, 1995; Barth et al., 1996;
Eccher et al., 1996; Nelson, 1996) and derivatives (Venkatachalam, 1996) held
by banks and insurance companies. Park et al. (1999) find value relevance of
recognized fair values for available-for-sale securities under SFAS 115. While
all these studies focus on financial sector firms, Simko (1999) for a cross-indus-
trial sample finds no significant sign of incremental value relevance for SFAS 107
disclosures, which is attributed to the insignificance of financial activities for
these firms. With respect to other financial instruments, notably loans held by
banks, results differ, which can be interpreted as lack of reliability due to
private information. On the other hand, Beaver and Venkatachalam (2000) find
value relevance for the discretionary component of loan fair values. The notion
of perceived insufficient reliability is especially critical for non-financial instru-
ments. Evidence so far rests on parallels from market-value regimes in Australia
and the UK and thus should be considered with caution. As an example, Barth
and Clinch (1998) find value relevance for the remeasurement differences of
non-current assets under Australian Generally Accepted Accounting Principles
(GAAP), yet further specification shows significant results only for negative
The Decision Usefulness of Fair Value Accounting 325
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.
As contribution to the theoretical literature, an economic analysis is undertaken,

which in contrast to previous research (see, e.g. Barth and Landsman, 1995;
Bromwich, 2004) is based explicitly on two specific approaches to decision use-
fulness, the measurement and the information perspective. Emphasis is put on the
evaluation of the paradigmatic foundations underlying regulators’ endorsement
of fair value measurement and on the comparative analysis of fair value account-
ing vs. historical cost accounting. Secondly, this paper proceeds to make a con-
tribution to the standard setting literature on the choice of measurement basis.
Resting on the precept of a decision-model perspective on accounting regulation,
potential implications of the analysis for questions standard setting bodies are
concerned with in their evaluation and implementation of fair value measurement
are discussed. The initiation of a common FASB/IASB conceptual framework
project and the recent publication of the IASB’s Discussion Paper on ‘Measure-
ment Bases for Financial Accounting’ indicate that there is demand for such
research.
One major result of this paper is that the conceptual foundations of the fair
value paradigm articulated on behalf of standard setters cannot be unequivocally
supported by theoretical reasoning. The validity of the paradigmatic foundations
appears particularly problematic for model-based estimation of fair value and
thus for valuation of non-financial positions. With regard to fair value as a
balance sheet measure (recognition), standard setters do not present any specific
case. Notably, no concept of fair value income is developed, despite the growing
use of fair value remeasurements. Application of different theoretical concepts of
informative income leads to varying perceptions of the usefulness of fair value
income and thus emphasizes the need to clarify and elaborate the concept of
fair value accounting prior to further implementation.
The remainder of the paper is organized as follows. Section 2 outlines the
concept and proliferation of fair value measurement in financial reporting stan-
dards, with special emphasis on the paradigmatic foundations. Section 3 develops
the methodology employed for the decision usefulness analysis of fair value
measurement. In Section 4, this methodology is applied to the analysis of fair
326 J.-M. Hitz
value per se (disaggregated reporting), while Section 5 in a comparative setting

explores the use of fair value for balance sheet and income measurement. Section
6 concludes with a summary of the basic results and a discussion of potential
implications for standard setting and future research.
2. Fair Value Accounting – A Shift in Standard Setting Paradigms

2.1. Fair Value: Definition and Estimation
Despite different wording, the definitions and meanings of the term ‘fair value’
are basically equivalent in FASB and IASB pronouncements. The general
FASB definition in SFAC No. 7 has recently been overhauled by the new Stan-
dard SFAS 157, Fair Value Measurements, which states in para. 5: ‘Fair value is
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 use of accepted, theoretically sound pricing methods, represents a technique

of last resort. Ample guidance exists on valuation models for financial instru-
ments, and accepted methods can be found in the marketplace. For non-financial
items, fair value estimation rests on a present value approach. SFAC No. 7, SFAS
157 and, with modifications, IAS 36, develop the principles and methods for such
measurements. Notably, they adopt an ‘economic’ view on measurement clearly
grounded in modern neo-classical finance theory, and distinguish traditional from
expected cash flow and residual earnings approaches.
To summarize, fair value represents a specific current value, that is, exit value
under idealized conditions. Estimation follows a three-tier process, with a strict
preference for market-based measures.
2.2. Paradigmatic Foundations

The move towards fair value measurement is frequently characterized as a shift
in paradigms (e.g. Barlev and Haddad, 2003, p. 383). We share the notion that
this process is based on firm beliefs and assumptions and thus deserves specific
attention, with evaluation addressing the theoretical underpinnings. Therefore,
the paradigmatic foundations of fair value measurement will be briefly
elaborated.
A paradigm can be defined as a set of values and beliefs shared by a
specific community.4 Accordingly, with respect to financial reporting, a para-
digm shall be defined as a set of shared beliefs on the objectives of financial
reporting and on the accounting principles by which these can be achieved. It
is grounded in elaborated assumptions, and characteristically requires a theoreti-
cal foundation or vindication. More specifically, a measurement paradigm rep-
resents a consensus on the measurement attributes required to achieve the
reporting objective in question. Once a financial reporting paradigm is adopted
by regulatory bodies, it becomes the guiding principle for accounting regulation,
that is, standard setting.
The fair value paradigm rests on the decision usefulness paradigm, which was
established as an official standard setting objective only with the formation of the
FASB and the conceptual framework project.5 Thus, while the historical cost
model emanated from a variety of influences, among them also aspects of con-
tracting and stewardship (Holthausen and Watts, 2001, p. 49), fair value measure-
ment has been introduced with explicit reference to a clearly stated reporting
objective: the provision of information to investors to enable them to assess
the amounts, timing and uncertainty of future cash flows from an investment in
a firm’s shares or debt securities (SFAC No. 1, para. 37; IASB Framework,
para. 15). Analysis of the relevant pronouncements identifies one theoretical
assumption that appears to constitute a fundamental pillar of the fair value para-
digm. According to this information aggregation hypothesis, the market price
aggregates in an efficient and virtually unbiased manner the consensus
328 J.-M. Hitz
expectations of investors in the market concerning the cash flow pattern of the
asset or liability:
An observed market price encompasses the consensus view of all market-

place participants about an asset or liability’s utility, future cash flows, the
uncertainties surrounding those cash flows, and the amount that market-
place participants demand for bearing those uncertainties.
(SFAC No. 7, para. 26)6
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
projections. Under such circumstances, market price information directly

satisfies the informational needs presumed for investors and therefore contributes
to financial reporting’s decision usefulness objective.
Standard setting bodies thus in an elaborate manner establish the conceptual
case for fair value measurement with reference to theoretical economic reasoning.
Naturally, the paradigmatic information aggregation hypothesis represents by no
means the sole driver behind the promotion of fair value measurement. There is a
wide range of other arguments, such as the comparability of market values and the
reliability of market price information. Plus, there are arguments specific to
certain positions, for example, the conceptual merits of market valuation for finan-
cial instruments and the accounting for risk management policies. Yet, the infor-
mation aggregation hypothesis is unique in its generality and theoretical backing.
Being regularly reaffirmed, it has guided the move towards fair value accounting
from its beginnings. Since information aggregation refers to the relevance of fair
value, that is, the correspondence of reported information and required infor-
mation, reliability concerns appear to be the prime argument capable of restricting
the implementation of fair value measurement in future standard setting projects,
especially where balance sheet recognition is concerned.
2.3. Evolution and Implementation of the Fair Value Paradigm

The move towards fair value measurement results from the adoption of the fair
value paradigm by standard setting bodies such as FASB and IASB. At the
roots of the development lies the perceived insufficiency of the then predominant
reporting model, the so-called revenue –expense approach, which is usually traced
to the Paton and Littleton (1940) monograph.8 Deficiencies of this approach had
been identified by the FASB already at the beginning of the 1980s and resulted in a
general overhaul of the accounting model. The move towards fair value account-
ing therefore fits into a more general development, the adoption of the so-called
asset-liability or balance sheet approach. The main property here is an income
definition that anchors on changes in assets and liabilities rather than on the
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-
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
For measurements at initial recognition or fresh-start measurements, fair

value provides the most complete and representationally faithful
330 J.-M. Hitz
measurement of the economic characteristics of an asset or a liability.

(SFAC No. 7, para. 36)
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.
2.4. Fair Value in Contemporary Accounting Standards

Given the gradual evolution of the fair value paradigm, its impact on current stan-
dards shall be summarized briefly. At present, both US GAAP and IFRS require
the disclosure of fair values for virtually all financial instruments (IFRS 7, SFAS
107). Guidance on fair value accounting for financial instruments is also identical
in principle. IAS 39 and SFAS 115, 133 require trading securities and derivatives
held for trading or as part of a fair value hedge to be measured at fair value with
revaluation gains and losses taken directly to income. Available-for-sale-
securities are also carried at fair value, but gains beyond the historical cost
ceiling are recognized as other comprehensive income until realization. This
recycling approach is also applied in accounting for derivatives that are part of
a cash flow hedge. In both regimes, securities classified as held-to-maturity,
non-securitized financial assets and obligations, except derivatives, are in prin-
ciple accounted for at cost. This mixed model approach reflects standard
setters’ reluctance and affected parties’ resistance to implementation of full
fair value accounting, despite the tentative consensus on its conceptual merits
especially on the relevance dimension (IASC, 1997; JWG, 2000). The IASB
has taken a big step in this direction with the 2003 revision of IAS 39, which
introduces the ‘fair value option’ to designate any financial instrument as
‘measured at fair value through profit and loss’ at inception. Objections
especially from bank regulators, notably the European Central Bank, resulted
in a partial endorsement by the EU only (‘carve out’) and prompted the IASB
to restrict the fair value option to areas where an accounting mismatch is elimi-
nated. With the recent publication of SFAS 159 in February 2007, the FASB
follows suit and implements a similar, yet less restrictive fair value option.
In contrast to the rules on accounting for financial instruments, US GAAP and
IFRS standards differ considerably in the use of fair value as a measurement attri-
bute for non-financial items. Notably, FASB standards at present require fair
value exclusively as a measure for impairment losses, that is, they invariably
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.
In summary, IFRS implement the fair value paradigm more progressively.

While the FASB obviously takes a cautious stance especially on revaluation of
non-financial items, the IASB adopts fair value measurement in a more consequent
manner, accepting the erosion of the twin pillars of the historical cost model, cost-
based measurement and transaction-based income recognition. Recent projects and
activities, for example, on accounting for insurance contracts, leases and business
combinations, illustrate the IASB’s commitment to fair value measurement, thus
underscoring the impetus and direction of the development.
2.4. Fair Value Debate and Research Questions

The implementation of the fair value paradigm has ever since been a contentious
issue. While most parties appear to agree on the benefits of fair value disclosures,
opinions differ substantially with regard to fair value measurement of recognized
items and the treatment of revaluation gains and losses. The fundamental ques-
tions surrounding the fair value debate can be summarized as follows (see, e.g.
Wyatt, 1991, p. 84; Barth, 2000, pp. 18 – 22):
. 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?
The aim of the following sections is to develop theoretical thoughts on the

decision usefulness of reporting fair value. In doing so, a standard setting
context is employed in order to discuss potential answers to these motivating
questions.
332 J.-M. Hitz
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.
The analysis is carried out as economic analysis and is unique in that it is

based on two conceptual viewpoints, the measurement and the information per-
spective, which the literature both associates with the increasing role of fair value
measurements.16
3. Methodology: Measurement and Information Perspectives and the

Economic Analysis of Financial Reporting Concepts
3.1. Measurement Perspective
The so-called measurement perspective represents the traditional view on the
information objective of financial reporting, especially of financial accounting.
It is rooted in the neoclassical theory of value and income developed by econom-
ists such as Hicks, Fisher and Lindahl (for an overview see Liang, 2001). The fun-
damental notion underlying the measurement perspective is that accounting
should directly measure and report the basic information required by investors,
which is the value of the firm, or at least a fraction of it. Thus, firm valuation
is delegated to the reporting entity. Under the measurement perspective, stocks
measures like assets, liabilities and equity and flows measures like income are
well defined and exhibit an economic character.
In an ideal world of complete and perfect markets, disclosure of the market
values for all the firm’s assets and liabilities directly reports firm value and
thus the desired investor information. Earnings equal economic income.
Obviously, the measurement perspective is embedded into such a scenario
(Beaver, 1998, pp. 4, 76; Barth, 2000, p. 15). Here, decision useful information
is information on the contribution of assets and liabilities to enterprise value.
Thus, the benchmark measurement attribute is value in use.
For a realistic setting, however, neither value nor income is a well-defined
concept and the rigid measurement perspective runs into difficulties (Beaver
and Demski, 1979). Yet, the measurement notion is influential as well for real-
world accounting, a fact witnessed by the traditional, unchallenged use of the
terminology of valuation in accounting (Beaver, 1998, p. 76; Barth, 2000, pp.

15 –18). Therefore, we distinguish the rigid measurement perspective from its
real-world corollary, the decision-model approach. Here, the decision problem
of a typical investor is regarded in order to directly delineate information
demands. For the purposes of this paper and in accordance with the FASB’s
and the IASB’s conceptual frameworks, the decision problem is reduced to secur-
ity valuation. Therefore, under this variant of the measurement perspective,
investors demand information that directly feeds into their present value calculus,
that is, decision useful information is (aggregated) information on the structure of
future cash flows. The use of such reported information is traded off against the
use of subjective cash flow estimates. Accordingly, the usefulness of reported
cash flow information depends on its descriptiveness, that is, its ‘quality’ and
cost –benefit considerations, both of which are dependent on the decision situ-
ation assumed for the typical investor.
3.2. Information Perspective

The measurement perspective represents one foundation of the earlier a priori
research (Liang, 2001, pp. 224 –229). Criticisms of the restrictive assumptions
underlying this view on ‘informative’ reporting were aggravated by the develop-
ment of information economics, which produced the influential ‘impossibility
result’ for normative accounting principles (Demski, 1973) and nourished the
establishment of a new research paradigm, the information perspective.
While the measurement perspective regards financial accounting numbers as
numerical inputs to security valuation models, the information perspective
takes a broader view (Vickrey, 1994, p. 1108; Demski et al., 2002, pp. 161 –
163). In information economics, useful information is defined in an abstract
manner as signals capable of transforming a priori expectations (beliefs) into a
posteriori expectations, which induces revisions and therefore improvements of
decisions. In the latter case, an information system has information content. If
the benefits of the improved decisions exceed the cost of information procure-
ment and processing, the information system also has information value. Com-
parisons of information systems can be conducted based on their fineness, that
is, their capability to partition the event space.
From an information perspective, financial reporting represents but one infor-
mation system competing with others.17 Since information is only relevant in its
capability to induce revisions of expectations, the presentation format does not
matter. Thus, in contrast to the measurement perspective, specific accounting rep-
resentations such as balance sheets, captions and categories such as assets, liabil-
ities, etc. are irrelevant.
The rise of the information perspective is conventionally associated with the
increasing focus on empirical accounting research (Beattie, 2002). Yet, infor-
mation perspective criteria can also be extracted and used for the purpose of
334 J.-M. Hitz
conceptual evaluation. For the purposes of this paper, two concepts of decision
usefulness from an information perspective shall be distinguished:
. Information content refers to the ‘newness’ of accounting information and is

assumed for such information that (1) is released for the first time to the
semi-strong form efficient stock market via financial reporting and (2) is
decision relevant, that is, capable of altering investors’ expectations with
respect to the value of the firm.
. Capital-markets-based research also recognizes a less rigid form of decision
usefulness: the function of financial statements to aggregate in an efficient
manner valuation-relevant information regardless of its timeliness, thus pro-
viding cost-efficient capital markets information (Barth, 2000, p. 16; Beaver,
2002, p. 461). The information aggregation function will therefore be con-
sidered as the second variant of decision useful information production

under the information perspective. It is assumed to apply when: (1) the data
in question would exhibit information content were they not known in
public (value relevance); and (2) the provision of these data via financial
reporting can be viewed as cost-efficient information aggregation. Since the
efficiency of information aggregation via financial reporting is outside the
scope of this paper, analysis of the information aggregation criterion will
focus on the value relevance of the information in question.
3.3. Methodological Background

Methodologically, the approach pursued in this paper represents a priori econ-
omic analysis, supplemented by standard setting inferences. Economic analysis
aims at identifying the contribution of financial reporting alternatives to societal
welfare or economic efficiency. Therefore, it is based on evaluative criteria that
can be reconstructed as hypothetical consensus of the majority of constituents
(Cushing, 1977).18 Neo-institutional economics and the theory of asymmetric
information provide a rich theoretical backing for the contracting and the valua-
tion function of financial reporting. The latter objective, the provision of infor-
mation useful for making investment decisions, is agreed to be a primary
factor in giving accounting societal value and therefore represents the basis for
our analysis. Measurement and information perspectives are concepts that are
established in the research literature as perspectives on this decision usefulness
objective. They can be conceptualized as conceivable consensus criteria.
One property of economic analysis is that it constitutes a priori research and
therefore produces results on hypothetical reporting alternatives prior to
implementation. Empirical research, on the contrary, characteristically represents
a posteriori research. Its applicability to questions of accounting regulation is thus
limited, especially with respect to predicting capital market reactions to new
accounting standards.19 Since the FASB’s and the IASB’s frameworks identify
decision usefulness as the primary objective of financial reporting, the findings
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
notion of decision usefulness needs to be clarified. In their frameworks, FASB

and IASB elaborate on the qualitative characteristics of decision useful infor-
mation. Primary characteristics are reliability and relevance. While reliability
in principle refers to the degree of intersubjective verifiability, relevance is
defined as the ‘capacity of information to make a difference in a decision by
helping users to form predictions about the outcomes of past, present, and
future events or to confirm or correct prior expectations’ (SFAC No. 2, Glossary
of Terms). Although this definition clearly has an information economics flavour
and standard setting bodies have been known to recognize results from empirical
capital-markets-based accounting research in their decisions, they have also on
numerous occasions resorted to measurement perspective arguments. More gen-
erally, though, the concept of relevance remains vague (Liang, 2001, p. 231).
This is in line with the complex welfare considerations and value judgments stan-
dard setters need to undertake. As a result, the conceptual frameworks of the
IASB and the FASB do not provide a theory in the sense that researchers and
others could predict their standard setting decisions (Barth et al., 2001, p. 90).
Thus, in accordance with the decision-model approach adopted here, evaluative
criteria derived from information and measurement perspectives represent but
conceivable views on decision usefulness from a standard setting perspective,
with no claim to correspondence. More specifically, since both perspectives
refer to the representational characteristics of accounting information and not
to its verifiability, they can be conceptualized as conceivable specifications of
the relevance criterion that standard setters would or would not share in the
light of circumstances.
3.4. Structure of the Analysis

The analysis proceeds as follows. The first part (Section 4) evaluates the decision
usefulness of the fair value measure on a stand-alone basis. That is, the reporting
format for fair value and its aggregation are not considered. This abstract per-
spective can be thought of as the full disclosure of the fair values for all the
336 J.-M. Hitz
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
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.
4. Decision Usefulness of Fair Value as a Measurement Attribute

4.1. Measurement Perspective
As a starting point, a rigid measurement perspective is adopted. In a world of
complete and perfect markets, fair value (FV) equals market value equals
value in use (VIU), the entity-specific value of a position (Barth and Landsman,
1995). The sum of fair values for all the firm’s assets and liabilities (N positions)
thus at any point in time (t) constitutes a precise measure of firm value (V ):
X
N X
N
Vt ¼ VIUnt ¼ FVnt :
n¼1 n¼1
Obviously, fair value represents an ideal, decision useful measurement attri-

bute under these conditions. Various proponents of fair value measurement, at
least implicitly, appear to have such a scenario in mind. Yet, this setting not
only represents an idealized world, it also does not have a role for financial
reporting: by definition, complete information is costlessly observed in the mar-
ketplace (Ronen, 1974, p. 147; Bromwich, 1977, p. 592). Financial reporting,
however, is an institution created by the deficiencies of real-world markets,

notably asymmetric information and transaction costs.
The measurement perspective in a real-world scenario of incomplete and
imperfect markets equally requires investors to agree on one concept of value.
Only if value is identically defined and therefore independent of individual pre-
ferences and beliefs can the valuation task be delegated to the financial reports,
that is, the reporting entity. Unanimity on the value concept prevails only when
consumption and investment decisions can be separated: in that case, investors
exclusively base their decisions on the present value of cash flows – no infor-
mation on the timing, amounts and uncertainty of future cash flows is required,
since investors establish their preferred cash flow and consumption patterns via
capital market transactions. Finance theory shows that such irrelevance of indi-
vidual preferences follows if markets satisfy the spanning and the competitivity
criterion (Grossman and Stiglitz, 1977; DeAngelo, 1981). Spanning prevails on a

well-developed capital market which allows cash flows from non-financial
investments to be duplicated (insured). Individual risk and timing preferences
are reflected in state prices which unambiguously determine value. The compe-
titivity assumption requires that neither investments in non-financial positions
nor capital market transactions have an impact on state prices. Spanning and
competitivity are thus required for the present value criterion (and the share-
holder value objective) to hold.
Real-world conditions can generally be expected to satisfy these separation cri-
teria in principle, except for ‘exotic’ investment projects that create cash flows
which cannot be hedged via capital market transactions. Still, value in use rep-
resents the benchmark measurement attribute from a measurement perspective.
However, fair value as a specific market value concept will normally not equal
value in use. Additionally, for many assets, especially for intangibles constituting
competitive advantages, no functioning markets exist. While firm value
conceptually equals the sum of the values in use for all identifiable assets and
liabilities, the respective fair valuation under realistic conditions systematically
underestimates firm value, because, unlike market value, a position’s value in
use incorporates two components: the asset or liability in question plus a fraction
of intangible assets, that is, management skill. On the firm value level, the sum of
these unidentifiable (not separable) intangibles equals goodwill (g), which cap-
tures the difference between firm value and market values of assets and liabilities:
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
leads to systematic undervaluation of a firm since market values do not incorpor-

ate competitive advantages resulting from specific intangible assets. Under these
circumstances, the rigid measurement perspective loses its conceptual appeal
(Beaver and Demski, 1979). Accordingly, the decision usefulness of fair value
measurement appears poor from this viewpoint.
Adopting the less restrictive decision-model approach, the question becomes
not whether fair value measurement can produce an unbiased measure of (frac-
tional) firm value, but whether it can improve individual valuations via the pro-
vision of (aggregated) information on the structure (distribution) of future cash
flows that directly feeds into the valuation model. Here, the case for potential
decision usefulness of fair value measurement can be made for activities which
are (1) not associated with rents and (2) do not interact with the firm’s other
activities and therefore can be separated for valuation purposes: investors can
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.
4.2. Information Perspective

From an information perspective, fair value’s contribution to decision usefulness
is not evaluated on the basis of its convergence with value in use, but on its capa-
bility to (1) alter expectations and thus revise decisions, or to (2) (efficiently)
aggregate value-relevant information.
Application of the rigid concept of useful information, information content,
produces a straightforward result: since fair value, by definition, is only to
include information publicly available in the marketplace, it cannot by itself

revise expectations of market participants. Disclosure of fair values thus has
no incremental information content, let alone information value. This is
especially true for fair values estimated via marking-to-market, that is, market
prices. Conceptually, this result also applies to synthetical fair values generated
by internal models, since the principle of market-based measurement requires the
use of publicly available market data and to emulate market expectations.
However, in practice internal estimates and assumptions, that is, private manage-
ment information, are incorporated into such fair values. This leads to the
awkward result that information content can only be generated where fair
value estimation violates the conceptual foundation of market-based measure-
ment. Of course, these results are of a rather theoretical nature, since fair
value measurement is applied to the entity-specific resources and obligations –
information which is inherently private and thus of potential information

content. Yet, the observation that for a scenario of full disclosure of an entity’s
assets and liabilities, full fair value measurement creates no incremental infor-
mation content because market participants can perform such a market valuation
themselves points at a certain contradictiveness of the fair value concept.
A comparatively stronger case for the usefulness of fair value measurement
can be made from the broad value relevance perspective, if fair value (in an effi-
cient manner) collects and aggregates value-relevant information. As pointed out,
value relevance is assessed based on the hypothetical question whether the
respective information were capable of altering investors beliefs and actions on
publication if it were not (yet) publicly available.
Therefore, for both the value relevance and the decision-model approach, the
ultimate evaluation of fair value rests on its informational properties – the ques-
tion of what kind of information it conveys and whether this information is of
valuation relevance/potential information content. This research question
obviously requires assessment of the theoretical validity of the information
aggregation hypothesis underlying the fair value paradigm. The next section
will address this issue, differentiating the two sources of fair value estimates,
market prices and marking-to-model.
4.3. Informational Properties of Fair Value

4.3.1. Marking-to-market
The interpretation of market price as the present value of future cash flows is well
accepted in economics and finance. Yet, it is not descriptive of the nature of the
expectations incorporated. More specifically, it is not clear what kind of infor-
mation, that is, what information set is processed and in what manner. Thus,
the analysis of the informational properties of fair value as a market price is inex-
tricably linked to the question of market efficiency. While the generic definitions
are attributed to the seminal work of Fama (1970), a more specific concept is used
for the purposes of this paper. The so-called Fama – Rubinstein efficiency
340 J.-M. Hitz
emphasizes the notion of ‘consensus expectations’ which is central to the fair

value paradigm. Accordingly, a market where naturally market participants
hold heterogeneous expectations is efficient with respect to one specific infor-
mation set. This information set can be conceptualized as consensus expectations,
the set of homogeneous expectations that, if held by all market participants,
would result in a price identical to the one witnessed in the presence of hetero-
genous expectations. Accordingly, prices evolve as if each investor held the iden-
tical information set, consensus expectations (Rubinstein, 1975, p. 818). This
concept of information efficiency illustrates that any market price can, in prin-
ciple, be viewed as an aggregate of consensus expectations (Verrecchia, 1979,
p. 960).
The relevant question now concerns the nature of this information set. The
assumption traditionally held in economics that market prices efficiently aggre-
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
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
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
financial reports, as laid down in standard setters’ frameworks. However, this

conditioning of expectations rests on strict assumptions, especially normal distri-
bution of cash flows, which are not descriptive of reality. Rather, one price can be
the result of an indefinite number of cash flow profiles. The notion of inferring
precise information on the timing, amounts and uncertainty of consensus cash
flow expectations therefore appears as not realistic, whereas the Bayesian revi-
sion of a subjective present value estimate certainly is descriptive of reality.
What are the implications of the mixed findings on the fair value paradigm’s
theoretical validity for the decision usefulness of market price reporting? The
result that market prices have information content in that their disclosure poten-
tially induces revisions of decisions exclusively based on individual information
sets demonstrates the valuation relevance of market price information. Applying
this result to aggregated reporting, the disclosure of a sum of market prices for
homogeneous positions, for example, fair values of trading securities, conveys

information relevant for the valuation of the firm as a whole. Investors learn
about the ‘consensus present value’ for the positions in question and can thus
extract useful information from financial reports. Therefore, given sufficient
information quality, the decision usefulness of (aggregated) market price disclos-
ures can be supported from the broad value relevance perspective. As pointed out,
the theoretical case for the incremental information content of disclosing publicly
available market prices is, on the other hand, weak, when disaggregated disclos-
ure of a firm’s assets and liabilities is assumed.
From a measurement, decision-model perspective, requirements for market
prices to represent useful information are more restrictive since the information
quality needs to be higher than or at least equal to the quality of the investors’
individual projections, which are substituted by market price information in a
separation calculus. Obviously, the decision to use a noisy aggregated consensus
forecast instead of one’s own projections is context-specific and critically
depends on the private information set. Yet, tentative reasoning suggests that
only prices for positions traded on highly liquid markets should be used, with
no evidence of investor sentiment or irrationally motivated biasedness. Similar
to the findings from the information perspective, the case for substitution
appears weak – the notion of improved estimation quality problematic – with
respect to prices for positions not traded on organized markets.
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
underlying the CAPM is perfect and complete markets, notably no transaction

costs and perfect information. Therefore, the valuation methods underlying fair
value modelling usually do not account for the influence of information
asymmetry on market pricing, which is of course due to the nascent state of
this line of research (O’Hara, 2003, p. 1336). Plus, these models assume equili-
brium states, while financial reporting is a result of disequilibrium situations
(Peasnell, 1977, p. 164). In brief, valuation methodology rests on strict assump-
tions not descriptive of reality that lead to expected systematic overvaluation of
assets vis-à-vis ‘real’ market prices.
A second fundamental informational feature of synthetic fair value is the lack
of verifiability and, as a consequence, of reliability. This is characteristic of any
economic valuation, which axiomatically rests on projections and expectations
about the future. Since such prospective data represents soft information, only
plausibility and consistency assessments can be made. Inevitably, this creates

opportunities for the exercise of management judgement and discretion and
thus for intentional bias (Benston et al., 2003, p. 39). Therefore, the well-estab-
lished argument that fair value measurement reduces incentives and opportunities
for management discretion appears to focus on market price valuation rather than
marking-to-model. It does not hold for the majority of non-financial assets which
cannot be marked to market and casts substantial doubt on the reliability of the
fair value concept.
However, the decision usefulness perspectives adopted for this analysis rep-
resent conceivable notions of the relevance criterion and refer to the nature
and amount of information aggregated in fair value estimates. A straightforward
result here is that model-based fair value cannot, by definition, represent an
aggregate of expectations dispersed in the marketplace: since valuation rests
on the information set of one person or one organization, this fair value loses
its capacity to efficiently collect and aggregate consensus expectations about
the cash flow profile of the relevant position. The paradigmatic information aggre-
gation assumption falters – a result that substantially questions standard setters’
theoretical reasoning on the desirability of fair value measurement. Rather than
market information, model-based fair value inevitably incorporates management’s
private information and assumptions, that is, elements of value in use.
From a measurement, decision-model perspective, this result forbids the qua-
lification of fair value as a sufficiently accurate aggregate of expected future cash
flows. The investor rather needs to trade off the benefits of private information
incorporation into such fair values (increased relevance) against the danger of
intentional bias (decreased reliability). Again, no clear-cut results emerge, yet
the overall suitability of such estimates as replacements for individual investor
expectations is reduced. The overall case for decision usefulness thus appears
to be relatively weak from a measurement perspective.
The collapse of the paradigmatic information aggregation assumption – the
impossibility to reproduce market expectations and thus to simulate ‘informative’
market prices – equally impairs the decision usefulness of synthetic fair value
344 J.-M. Hitz
from an information perspective. Reconstruction of fair value measurement as

(efficient) aggregation of value-relevant information fails. Curiously, model-
based fair value is capable of creating useful information in the strict, information
content sense, when credible communication of private management information
takes place. Empirical evidence suggests this is happening (Barth and Clinch,
1998; Beaver and Venkatachalam, 2000). Accordingly, fair value measurement
on the marking-to-model level receives vindication only where the fair value
definition is violated and elements of value in use are incorporated. Obviously,
this observation refutes rather than confirms the theoretical basis of fair value
measurement.
As a conclusion of the first part of this analysis, the decision usefulness of dis-
aggregated fair value information can be justified from an information and from a
measurement perspective, yet only under specific conditions. Notably, the para-
digmatic foundations of fair value measurement appear theoretically valid only

for prices taken from organized, sufficiently liquid markets and can therefore
hardly be applied to reporting for non-financial items, which typically require
model-based estimation. Where market prices are used, additional concerns
arise because such measurement of assets and liabilities is based on publicly
available information not specific to the entity. In any case, the general claim
of the paradigmatic information aggregation hypothesis appears as too ambitious.
Comparative aspects have not been considered in this section since historical
cost as a measurement attribute does not lend itself to a stand-alone analysis from
the perspectives adopted here. Theoretical justifications of the use of cost-based
measures do not rest on informational properties such as cash flow indication.
Putting aside these caveats, it is fairly straightforward that cost represents an esti-
mate of fair value at inception. Yet, at subsequent measurement dates, the equiv-
alence to notions of economic value loosens up due to the realization and
matching principle. Notably, accounting recognition of value changes occurs
in a biased way and is deferred in time, which hints at comparatively weaker
inference of market expectations from such carrying amounts (book values).
Yet, as pointed out, historical cost accounting is primarily aimed at income deter-
mination. Therefore, the purpose of the next section is to look at the reporting
issues involved, that is, to investigate aspects of the form of reporting fair
values, with an emphasis on the comparative evaluation of income concepts.
5. Decision Usefulness of Fair Value Accounting

5.1. The Case for Fair Value Accounting
From a strict information perspective, the form of financial reporting is irrelevant.
Thus, by addressing accounting recognition, we assume that issues such as rec-
ognition vs. disclosure do indeed matter. The relevance of the presentation
format may be due to cost of information acquisition and processing, or contract-
ing uses of accounting data which are not addressed here. The empirical evidence
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
information and measurement perspective so far adopted will serve as a frame-

work. Despite articulation, we distinguish between balance sheet valuation and
fair value income for expository reasons.
5.2. Fair Value Balance Sheet

The move from historical cost accounting towards fair value accounting reflects
an increasingly economics-based approach to accounting measurement, where
economic values are reported on the balance sheet. This emphasis on the valua-
tion function of the balance sheet, coupled with the paradigmatic foundations that
rest on a stocks perspective, illustrates the growing importance of the balance
sheet as a stand-alone instrument of investor information (Razaee and Lee,
1995, p. 217; Barker, 2004, p. 166). In the terminology of accounting theory,
the asset-liability approach is emphasized vis-à-vis the traditional revenue–
expense approach.
Since the information perspective does not allow for evaluation of different
reporting formats, its sole implication here is that, in accordance with the
results of the previous sections, the high degree of aggregation associated with
balance sheet format reporting substantially inhibits the inference of the under-
lying consensus present values, let alone the cash flow profiles. This is especially
so where both fair values based on market prices and model-based estimates are
aggregated and, more severely, where heterogeneous positions valued at different
measurement attributes are summed up in one balance sheet caption.21 From a
strict information perspective, the high degree of information aggregation
which is characteristic of balance sheet measures thus leads to a negative assess-
ment of potential decision usefulness.
The prior results from the measurement perspective evaluation of fair valua-
tion are independent of aggregation and can therefore be applied to balance
sheet measurement. Unlike historical cost accounting, fair value measurement
eliminates hidden reserves for recognized assets and thus narrows the gap
between accounting book value and enterprise value (market value of equity).
346 J.-M. Hitz
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).
Thus, it confirms the valuation relevance of aggregated accounting data.22

Some proponents even argue the superiority of accounting-based valuation,
since it relieves the valuation task of the difficult forecasting of dividends or
free cash flows, reduces the fraction of firm value to be estimated based on pro-
jections, notably the terminal value problem, and thus ‘brings value forward in
time’ (Penman, 1998, 2003, p. 160; Penman and Sougiannis, 1998). Although
accounting valuation theory is silent on the comparative advantages of different
measurement attributes, this basic idea suggests that fair value accounting has
advantages compared to historical cost accounting, since it ‘brings more value
on the balance sheet’ and ceteris paribus reduces the present value of residual
earnings to be forecasted. Expected fair value earnings are computed as required
return times opening assets and lend themselves to a straightforward interpretation
as the normal market return expected from employing the entity’s net assets. Of
course, the case for the relative desirability of fair value accounting from a
decision-model approach rests on rather crude reasoning and requires further
elaboration. Apparently, the predictive ability of earnings and residual earnings
under different measurement regimes is crucial and requires closer evaluation.
To summarize, while the strict measurement perspective on fair value as a
balance sheet measurement attribute reinforces the general result that fair
value is but an incomplete measure of partial firm value, intuitive reasoning
suggests superior decision usefulness of fair value accounting from a decision-
model (measurement) perspective. The increasing economic character of the
balance sheet reduces the valuation gap and focuses the estimation problem on
future (residual) earnings. The informational properties of fair value income
vs. historical cost earnings are therefore the subject of the following section.
5.3. Fair Value Income

5.3.1. On the evaluation of income concepts
The discussion of fair value income in the literature and by regulators so far
appears somewhat paradoxical. The properties and presumed decision usefulness
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
volatility of earnings, with negative implications for predictive ability (e.g.

Christie, 1992, p. 87; Wilson and Rasch, 1998, p. 24; Poon, 2004, p. 40). In a
more subtle vein, it is argued that restricting fair value measurement to certain pos-
itions will lead to ‘artificial’ volatility and ‘distorted’ earnings, because revaluation
gains and losses are economically compensated by the valuation differences for
positions not measured at fair value (‘mismatching’) (e.g. Beatty, 1995, p. 28;
Mauriello and Erickson, 1995, p. 181). This argument was first discussed with
the introduction of SFAS 115, which does not allow fair valuation of financial
liabilities, yet in a compromise solution alleviates volatility concerns by reporting
fair value differences for available-for-sale securities outside earnings.
Proponents of fair value income, on the other hand, stress its economic char-
acter and argue that remeasurement gains and losses reflect real economic vola-
tility (e.g. Sprouse, 1987, p. 88; Wyatt, 1991, p. 84). Plus, income realization
based on ‘objective’ market values allegedly deprives management of a device
for earnings management (e.g. Barlev and Haddad, 2003, pp. 384, 395).
Inherent in the volatility debate, which serves as a motivation for further analy-
sis, are at least two problems. First, the notion that the move towards fair value
accounting leads to increased earnings volatility appears to be an undue general-
ization. It rests on an isolated view on a single position and neglects compen-
sation effects when a number of positions, especially both assets and liabilities,
are remeasured at fair value. Therefore, volatility implications need to be con-
sidered separately for each proposed rule or standard. Second, evaluation of
fair value vis-à-vis historical cost concepts of income requires a notion of ‘appro-
priate volatility’, that is, a concept of informative income (Gellein, 1986, p. 16).
Yet, due to the impossibility theorem, there is no single, universally accepted
definition of income under conditions of imperfect and incomplete markets
(Beaver and Demski, 1979). More problematically, standard setting bodies do
not take on their role, which is to make such welfare-relevant decisions:
neither FASB’s nor IASB’s framework exhibits a clear definition of the
income concept pursued, despite an emphasis on the informational role of earn-
ings (Barker, 2004, pp. 158, 164).
348 J.-M. Hitz
With this background, a comparative analysis of fair value income is con-

ducted. In line with the methodology adopted here, measurement and information
perspective frame the analysis. While economic income is the earnings concept
implied by a rigid measurement perspective (Beaver, 1998, pp. 49, 57, 64– 67),
real-world circumstances demand predictive ability of accounting income, which
for the purposes of this paper is operationalized through the evaluative criterion
of earnings persistence. Thus, economic income and persistent earnings are ident-
ified as two conceivable views on decision useful income measurement. In a
comparative setting, they are applied to the theoretical scenarios of pure fair
value accounting (event-based income) and pure historical cost accounting
(transaction-based income).
5.3.2. Economic income (measurement perspective)

Economic income is usually associated with Hicks’ definition of income, accord-
ing to which ‘a person’s income is what he can consume during the week and still
expect to be as well off at the end of the week as he was at the beginning’ (Hicks,
1946, p. 176). It is the change in firm value during one period and thus a direct
measure of individual welfare or consumption potential. In a setting of certainty,
economic income equals the interest on firm value at the beginning of the period.
Given uncertainty, an unexpected component is added to this expected income.
Because economic income requires an unambiguous definition of value, it is
well defined only in a neoclassical setting. Yet, it is of conceptual merit for evalu-
ation of real-world income concepts as well, since it allows for comparative
analysis of relevant properties and emphasizes differences compared to ideal
‘earnings quality’ (Schipper and Vincent, 2003, p. 97). Therefore, some obser-
vations shall be made with respect to the conjecture that fair value income rep-
resents a superior estimate of economic income since it rests on economic
valuation (i.e. market valuation). In doing so, we assume that rents, that is, net
present value, are a component of economic income.23
Any concept of accounting income that satisfies the clean surplus condition
will over time result in accumulated earnings that equal the cash flow surplus.
Thus, over the life cycle of a firm, fair value income equals historical cost
income equals economic income. Differences are only inter-temporal and
result from different degrees of delayed or biased recognition of accounting
income. For fair value income, recognition is less delayed, that is, the gap
between creation and accounting recognition of value is smaller than for histori-
cal cost accounting, since the realization principle and the cost ceiling are aban-
doned. Income realization does not anchor on market transactions but on market
valuation and is thus directly linked to economic events. Yet, even under fair
value accounting, value differences attributable to unrecognized intangibles
and goodwill are not recognized until they result in cash transactions. Fair
value income as one variant of an economic approach to income measurement
differs systematically from the concept of Hicksian income. This is especially
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
express economic reality. Therefore, from an economic income perspective,

earnings variations even in a full fair value model incorporate elements of ‘arti-
ficial volatility’.
5.3.3. Earnings persistence

The economic analysis approach pursued here dispenses with the impossibility
result for income measurement by adopting conceivable consensus criteria for
informative income (Ohlson, 1987, pp. 2, 6). While economic income already
incorporates firm valuation and thus does not leave a prediction task for investors,
the prime concern of investors in a realistic setting is the formation of expec-
tations about the future. Therefore, the concept of predictive ability can be
seen as one conceivable consensus criterion for decision usefulness (Beaver
et al., 1968). It can be reconciled both with information and measurement per-
spectives on decision usefulness, yet it can also be regarded as a self-sufficient
evaluative perspective on informative income. In accordance with notions
hinted at in the frameworks,24 above all for its prevalence in valuation practice
and accounting theory (Beaver, 1999, p. 163; Schipper and Vincent, 2003,
p. 99), predictive value shall be measured by earnings persistence.
Earnings persistence refers to the rate at which present earnings persist into
future periods, that is, the degree to which a change in present earnings will be
reproduced in the next reporting period and, conceivably in a declining
pattern, for periods thereafter. The concept thus assumes that investors conduct
valuations based on estimates of future earnings. The residual income model pro-
vides one rationale for this assumption. In a more precise specification, earnings
persistence is defined by the autocorrelation of unexpected earnings components
(Lipe, 1990, pp. 50, 52). As two extreme points of theoretical reference, perma-
nent and transitory earnings can be distinguished. While a permanent earnings
shock is assumed to persist in equal amount throughout all future periods, tran-
sitory earnings shocks have no implications for future earnings at all (zero auto-
correlation) (Ramakrishnan and Thomas, 1998; Ohlson, 1999).
350 J.-M. Hitz
Opponents of fair value accounting who resort to the volatility criticism,

according to which introduction of fair value accounting creates erratic earnings
movements that deprive accounting income of its predictive ability, at least
implicitly adopt a persistent earnings income concept. This notion in many
cases stems from the traditional conjecture in accounting theory that focusing
on either income statement or balance sheet results in a decline of the information
quality of the opposite reporting instrument. Thus, it is argued, the focus of fair
value accounting on balance sheet information comes at the price of a loss in the
informativeness of the income statement (e.g. Christie, 1992, pp. 95, 101). The
validity of this conjecture deserves closer examination, starting on the level of
single positions measured at fair value.
Persistence of changes in fair value. The assumption of erratic, non-persistent

moves in asset values is a traditional subject of debate in the literature on efficient
(capital) markets. Notably, the characterization of asset values following a
Martingale process appears to lend vindication to the conjecture of decreasing
earnings persistence. Values (prices) follow a Martingale process if today’s
value represents the best estimate, that is, expected value, of the future value
(Et [Vtþ1] ¼ Vt). The well-known random walk property satisfies this definition
(LeRoy, 1989, p. 1589). For an asset following a Martingale, the fair game prop-
erty of efficient markets theory applies. It states that for a given information set,
no abnormal return can be made on an efficient market by using that information.
More specifically, with respect to the fair value concept, which rests on the
assumption of semi-strong form efficiency, the expected abnormal return from
using publicly available information for asset valuation is zero. Any deviation
from expected value, that is, current value, is therefore by definition unexpected,
and there is an equal chance of positive and negative deviations. This is the result
of the Samuelson theorem, which on the surface appears to present a case against
the usefulness of fair value income from an earnings persistence view: since
today’s value incorporates all available information, any deviation from it
cannot be predicted (Samuelson, 1965, 1973). Differences between expected
return and realized return, that is, ‘unexpected fair value income’, are unpredict-
able and uncorrelated, thus purely transitory.
However, a closer look shows that the Martingale property cannot be general-
ized for any market value or market price, since the fair game property refers to
the return on an asset, which consists of both a revaluation and a cash flow com-
ponent. Only where the expected earnings (market return) on the asset
(k FVt21) equal the expected cash flow (Et21 (ct)) will any deviation from
today’s value be transitory, that is, unexpected, because:
FVt FVt1 ¼ ½k FVt1 Et1 (ct ) þ ½FVt Et1 (FVt ) :

|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl} |fflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflffl}
expected change in fair value fair value shock
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}
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,
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
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
incorporates more information than transaction-based income, whose stability

may be deceptive. From such a perspective, the persistence of historical cost
income may appear far more ‘artificial’ than the ‘volatility’ of fair value income.
Our analysis of fair value income illustrates that the positions taken and con-
jectures made on its desirability rest on specific notions of informative income,
with different implications, for instance, for the interpretation and evaluation
of earnings volatility. Numerous arguments against and in favour of fair value
accounting cannot be compared because the underlying notions of informative
income are incommensurable. Final conclusions and evaluations of desirability
thus cannot be made. This discussion points at a fundamental insufficiency under-
lying the debate and the shift to fair value accounting: up to the present day,
accounting standard setters fail to communicate the income concept that is
pursued with the implementation of fair value measurement. Discussion so far
suggests that they have none, which severely inhibits future progress and consist-
ency in standard setting. The results here indicate that the relevant concept is not
one of earnings persistence, which appears to serve as a problematic vindication
for the transaction-based model (revenue –expense approach). Rather, the infor-
mation content approach used here in an intuitive fashion suggests that with
regard to the transportation of valuation-relevant information, fair value
income represents the conceptually superior approach, provided that fair value
shocks are separately disclosed.
One implication of this analysis for future research is that the informational
properties and time-series behaviour of residual fair value income deserve
deeper exploration. Coupled with the merits of ‘bringing value forward in
time’, one advantage of fair value (residual) income is that it focuses on econ-
omic rents and is less disturbed by effects of delayed recognition.
6. Conclusions and Standard Setting Implications

This paper analyses the potential decision usefulness of the fair value measure-
ment attribute from a theoretical viewpoint, based on the measurement and the
354 J.-M. Hitz
information perspective, and evaluates the results in a standard setting context.

As a point of reference, the paradigmatic assumptions underlying the move to
market valuation are extracted from standard setters’ pronouncements.
The analysis of the fair value measurement attribute supports the decision use-
fulness for fair value as a price taken from liquid markets. The conceptual case
for marking-to-model, on the other hand, is less strong. Notably, the paradigmatic
assumptions do not hold for the fair value of most non-financial assets – an obser-
vation which goes to the core of standard setters’ theoretical backing for fair
value measurement and its general claim to desirable informational properties.
In a second step, fair value accounting is analysed. Standard setters offer no
specific theoretical reasoning on the desirability and implementation of fair
value as a balance sheet and income measure. More strikingly, no income
concept is outlined. Application of an established measure of predictive ability,
earnings persistence, suggests the relative inferiority of fair value income

vis-à-vis transaction-based income and therefore lends support to criticisms of
unrepresentative income volatility. On the other hand, informational analysis
hints that, disaggregated appropriately, fair value income is a superior indicator
of the occurrence and persistence of valuation-relevant economic events, sup-
porting the informational role of such earnings volatility. The merits of fair
value income cannot be ultimately resolved without value judgements as to the
notion of informative earnings. However, since fair value measurement puts
more emphasis on balance sheet valuation and implies a more economic
concept of income, the properties of residual fair value income appear as a worth-
while path for further investigation.
These conclusions are based on the specific perspectives on decision useful-
ness adopted for the purposes of this analysis. Naturally, these perspectives
need not accord to standard setters’ evaluative criteria and their implicit value
judgements. However, if IASB and FASB paid attention to these views, this
research were to carry at least three tentative standard setting implications:
(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
theory is offered for generalizing the fair value paradigm to non-financial

items such as property, plant and equipment, or even intangibles.
(3) The special case against incorporating fair value measures into the core
financial statements receives additional support from the vagueness of the
income concept thus pursued. Theoretical reasoning demonstrates that the
relative superiority of fair value income over transaction-based income
varies critically with the notion of predictive ability applied. As long as stan-
dard setters are hesitant to elaborate on their notion of fair value income and
its contribution to decision usefulness, the transaction-based income concept
should be sustained for non-financial items.
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-
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
acknowledged. Finally, participation in the European Accounting Association’s

2002 doctoral colloquium provided useful comments at a very early stage of
this project, and lent valuable stimulus to my research.
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-
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).
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
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
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.
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.
Canning, J. (1929) The Economics of Accountancy (New York: Ronald Press).

Chambers, R. J. (1965) Measurement in accounting, Journal of Accounting Research, 3(1), pp. 32–62.
Christensen, J. A. and Demski, J. S. (2003) Accounting Theory: An Information Content Perspective
(Boston: McGraw-Hill).
Christenson, C. (1983) The methodology of positive accounting, Accounting Review, 58(1), pp. 1–22.
Christie, A. A. (1992) An analysis of the properties of fair (market) value accounting, in: K. Lehn and
R. W. Kamphuis (Eds) Modernizing U.S. Securities Regulation: Economic and Legal Perspec-
tive, pp. 85– 109 (New York: Irwin).
Cole, C. (1992) Moving toward market value accounting, Journal of Corporate Accounting and
Finance, 3(3), pp. 537–544.
Cushing, B. E. (1977) On the possibility of optimal accounting principles, Accounting Review, 52(2),
pp. 308–321.
DeAngelo, H. (1981) Competition and unanimity, American Economic Review, 71(1), pp. 18–27.
Demski, J. S. (1973) The general impossibility of normative accounting standards, Accounting
Review, 48(4), pp. 718– 723.
Demski, J. S., Fellingham, J. C., Ijiri, Y. and Sunder, S. (2002) Some thoughts on the intellectual foun-
dations of accounting, Accounting Horizons, 16(2), pp. 157 –168.
Diamond, D. W. and Verrecchia, R. E. (1981) Information aggregation in a noisy rational expectations
economy, Journal of Financial Economics, 9, pp. 221– 235.
Eccher, E. A., Ramesh, K. and Thiagarajan, S. R. (1996) Fair value disclosures by bank holding
companies, Journal of Accounting and Economics, 22, pp. 79 –117.
Edwards, E. O. and Bell, P. W. (1964) The Theory and Measurement of Business Income (Berkeley:
University of California Press).
Fama, E. F. (1970) Efficient capital markets: a review of theory and empirical work, Journal of
Finance, 25(2), pp. 383– 417.
Feltham, G. A. and Ohlson, J. A. (1995) Valuation and clean surplus accounting for operating and
financial activities, Contemporary Accounting Research, 11(2), pp. 689 –731.
Gellein, O. S. (1986) Financial reporting: the state of standard setting, in: B. N. Schwartz (Ed.)
Advances in Accounting, pp. 3 –23 (Greenwich, CT: JAI Press).
Gellein, O. S. (1992) Primacy: assets or income?, in: G. J. Previts (Ed.) Research in Accounting Regu-
lation, Vol. 6, pp. 197– 199 (Greenwich, CT and London: JAI Press).
Glosten, L. R. and Milgrom, P. R. (1985) Bid, ask and transaction prices in a specialist market with
heterogeneously informed traders, Journal of Financial Economics, 14, pp. 71–100.
Grossman, S. J. (1976) On the efficiency of competitive stock markets where traders have diverse
information, Journal of Finance, 31(2), pp. 573 –585.
Grossman, S. J. (1981) An introduction to the theory of rational expectations under asymmetric infor-
mation, Review of Economic Studies, 48, pp. 541 –559.
360 J.-M. Hitz
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.
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.
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
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
concepts, Journal of Business Finance and Accounting, 21(8), pp. 1097–1110.

Vives, X. (1995) The speed of information revelation in a financial market mechanism, Journal of
Economic Theory, 67, pp. 178–204.
Walker, R. G. (1992) The SEC’s ban on upward asset revaluations and the disclosure of current values,
Abacus, 28(1), pp. 3–35.
Watts, R. L. (2003) Conservatism in accounting – Part I: Explanations and implications, Accounting
Horizons, 17(3), pp. 207– 221.
Watts, R. L. and Zimmerman, J. L. (1990) Positive accounting theory: a ten year perspective, Account-
ing Review, 65(1), pp. 131–156.
White, L. J. (2003) The savings and loan debacle: a perspective from the early twenty-first century,
New York University Salomon Center, Working Paper Series, Nr. S-03-1.
Wilson, A. C. and Rasch, R. H. (1998) New accounting for derivatives and hedging activities, CPA
Journal, 68, pp. 22–27.
Wyatt, A. (1991) The SEC says: mark to market!, Accounting Horizons, 5(1), pp. 80–84.
Zhang, X. (2000) Conservative accounting and equity valuation, Journal of Accounting and Econ-
omics, 29(1), pp. 125–149.

Pertemuan 7-2007-Hitz-The Decision Usefulness of FairValue Accounting - A Theoretical

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pertemuan 7-2007-Hitz-The Decision Usefulness of FairValue Accounting - A Theoretical

Uploaded by

Copyright:

Available Formats

This article was downloaded by: [University of Stellenbosch]

On: 01 May 2013, At: 15:41

European Accounting Review

The Decision Usefulness of Fair

To link to this article: http://dx.doi.org/10.1080/09638180701390974

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-

The Decision Usefulness of Fair

University of Cologne, Germany

0963-8180 Print/1468-4497 Online/07/020323–40 # 2007 European Accounting Association

importance of fair value as an accounting measurement attribute. Since the mid-

market values to incorporate, in an efficient and virtually unbiased manner,

As contribution to the theoretical literature, an economic analysis is undertaken,

value per se (disaggregated reporting), while Section 5 in a comparative setting

2. Fair Value Accounting – A Shift in Standard Setting Paradigms

the use of accepted, theoretically sound pricing methods, represents a technique

2.2. Paradigmatic Foundations

An observed market price encompasses the consensus view of all market-

projections. Under such circumstances, market price information directly

2.3. Evolution and Implementation of the Fair Value Paradigm

For measurements at initial recognition or fresh-start measurements, fair

measurement of the economic characteristics of an asset or a liability.

2.4. Fair Value in Contemporary Accounting Standards

In summary, IFRS implement the fair value paradigm more progressively.

2.4. Fair Value Debate and Research Questions

The aim of the following sections is to develop theoretical thoughts on the

The analysis is carried out as economic analysis and is unique in that it is

3. Methodology: Measurement and Information Perspectives and the

terminology of valuation in accounting (Beaver, 1998, p. 76; Barth, 2000, pp.

ation assumed for the typical investor.

3.2. Information Perspective

. Information content refers to the ‘newness’ of accounting information and is

sidered as the second variant of decision useful information production

3.3. Methodological Background

notion of decision usefulness needs to be clarified. In their frameworks, FASB

3.4. Structure of the Analysis

4. Decision Usefulness of Fair Value as a Measurement Attribute

Obviously, fair value represents an ideal, decision useful measurement attri-

however, is an institution created by the deficiencies of real-world markets,

criterion (Grossman and Stiglitz, 1977; DeAngelo, 1981). Spanning prevails on a

leads to systematic undervaluation of a firm since market values do not incorpor-

4.2. Information Perspective

include information publicly available in the marketplace, it cannot by itself

information which is inherently private and thus of potential information

4.3. Informational Properties of Fair Value

emphasizes the notion of ‘consensus expectations’ which is central to the fair

financial reports, as laid down in standard setters’ frameworks. However, this

homogeneous positions, for example, fair values of trading securities, conveys

underlying the CAPM is perfect and complete markets, notably no transaction

plausibility and consistency assessments can be made. Inevitably, this creates

from an information perspective. Reconstruction of fair value measurement as

digmatic foundations of fair value measurement appear theoretically valid only

5. Decision Usefulness of Fair Value Accounting

information and measurement perspective so far adopted will serve as a frame-

5.2. Fair Value Balance Sheet

Thus, it confirms the valuation relevance of aggregated accounting data.22

5.3. Fair Value Income

volatility of earnings, with negative implications for predictive ability (e.g.

With this background, a comparative analysis of fair value income is con-

5.3.2. Economic income (measurement perspective)

express economic reality. Therefore, from an economic income perspective,

5.3.3. Earnings persistence

Opponents of fair value accounting who resort to the volatility criticism,