Description
Anderson,Brown,Hodder,andHopkins(ABHH,inpress)provideexperimentalevidencethatinvestorsarebetterabletoassessmanagementstewardshipandfirmvaluewhenfinancialstatementshighlightfir-valueinformationratherthanamortized-costinformation.Wefocusourdiscussiononfivemethod-ologicalchoicesthattheauthorsmake.WeconsiderhowthosechoicesaffectthegeneralityofABHH’sresults,consideranalternativetheoreticalexplanationforthoseresults,andsuggestopportunitiesforfutureresearchthatbuildonABHH’swork.
Accounting, Organizations and Society 46 (2015) 115–118
Contents lists available at ScienceDirect
Accounting, Organizations and Society
journal homepage: www.elsevier.com/locate/aos
Discussion of “The effect of alternative accounting measurement bases
on investors’ assessments of managers’ stewardship”
R
Scott A. Emett, Mark W. Nelson
?
S.C. Johnson Graduate School of Management, Cornell University, United States
a r t i c l e i n f o
Article history:
Available online 29 August 2015
a b s t r a c t
Anderson, Brown, Hodder, and Hopkins (ABHH, in press) provide experimental evidence that investors
are better able to assess management stewardship and ?rm value when ?nancial statements highlight
fair-value information rather than amortized-cost information. We focus our discussion on ?ve method-
ological choices that the authors make. We consider how those choices affect the generality of ABHH’s
results, consider an alternative theoretical explanation for those results, and suggest opportunities for
future research that build on ABHH’s work.
© 2015 Elsevier Ltd. All rights reserved.
1. Introduction
Anderson, Brown, Hodder, and Hopkins (hereafter ABHH, in
press) seek to inform a longstanding debate about the extent to
which fair value and amortized cost measurement approaches aid
?nancial statement users in assessing stewardship and in mak-
ing investment decisions. As argued by ABHH and others (e.g., see
Hodder, Hopkins, & Schipper, in press), the accounting literature
often assumes (but does not offer much evidence) that histori-
cal amortized-cost-based ?nancial information is inherently supe-
rior for stewardship purposes relative to fair-value-based ?nancial
information. ABHH provide evidence that fair value information
yields better stewardship assessments than does amortized cost in-
formation in their experimental setting, informing this debate.
One way to view ABHH’s contribution is that they invalidate
the assumption that amortized-cost information is always superior
to fair value information for assessing stewardship by demonstrat-
ing an instance in which the converse is true. Their work should
naturally lead to further research considering when each measure-
ment attribute (amortized cost or fair value) is more helpful for
assessing stewardship and/or providing decision-useful information
to investors. Therefore, we focus our discussion on ?ve of ABHH’s
methodological choices that likely affect the generality of their re-
R
We thank the authors of Anderson et al. (2015) (Spencer Anderson, Jason Brown,
Leslie Hodder, and Pat Hopkins), as well as Rob Bloom?eld and Bob Libby, for pro-
viding comments on this discussion. We gratefully acknowledge Deloitte & Touche
LLP’s ?nancial support of the AOS Estimates Conference.
?
Corresponding author.
E-mail address: [email protected] (M.W. Nelson).
sults and that offer opportunities for future research that modi?es
these choices. We also discuss an alternative theoretical explana-
tion for ABHH’s results which future research could seek to differ-
entiate from the explanation that ABHH provide.
2. Methodological choices and generality of results
2.1. Choice 1: De?ning good stewardship as actions that maximize
the present value of future cash ?ows
De?ning stewardship requires specifying the rights and respon-
sibilities of an agent in relation to a principal (Bushman, Engel, &
Smith, 2006; Bushman & Indjejikian, 1993; Chen, 1975; Cormier,
Magnan, & Morard, 2000; Gjesdal, 1981; O’Connell, 2007). ABHH
create a setting in which participants assume the role of investor
and receive higher payoffs by investing more in (or assessing
higher manager quality of) ?rms that increase the most in ?rm
value. In this setting, managers have clear rights – the right to
use the investor’s capital – and clear responsibilities – the respon-
sibility to increase ?rm value. ABHH’s de?nition of stewardship
matches this setting: they de?ne stewardship as “the actions of
managers in the discharge of their responsibilities to preserve the
value of investors’ capital investment and to earn a commensurate
return on their investment” (p. 1–2). This de?nition links stew-
ardship directly to ?rm value: decisions that increase ?rm value
unequivocally imply good stewardship, and decisions that decrease
?rm value unequivocally imply poor stewardship. A measurement
basis that is useful to investors in assessing stewardship in this
setting should always similarly be useful in making investment al-http://dx.doi.org/10.1016/j.aos.2015.08.003
0361-3682/© 2015 Elsevier Ltd. All rights reserved.
116 S.A. Emett, M.W. Nelson/ Accounting, Organizations and Society 46 (2015) 115–118
location decisions. Thus, ABHH’s study does not inform us about
how stewardship assessments might differ from ?rm valuation
assessments.
Different settings might operationalize different notions of
stewardship and produce conclusions from a stewardship perspec-
tive that differ from those of a ?rm valuation perspective. For ex-
ample, Watts (2003) argues that ?nancial accounting has evolved
to be conservative because many parties that contract with the
?rm have asymmetric payoffs from their contracts. Those parties
are concerned about the ?rm’s ability to reach some threshold
level of pro?tability, but are less concerned about the extent to
which the ?rm exceeds that threshold of pro?tability. Such parties
might view a manager as being a better steward if the manager
invests in lower-risk projects after meeting a threshold of prof-
itability, even if that means turning down high risk projects with
a positive net present value that would enhance ?rm value. Sim-
ilarly, investors for whom company-speci?c risks are undiversi?-
able might prefer manager behaviors that are more conservative
than might otherwise be preferred. These examples imply a de?ni-
tion of stewardship that emphasizes capital preservation, such that
some decisions that increase expected ?rm value may imply poor
stewardship, and some decisions that decrease expected ?rm value
may imply good stewardship. Future research could examine these
settings in order to operationalize other notions of stewardship.
2.2. Choice 2: Investors know that demand shocks are completely
exogenous
One concern with measuring assets and liabilities at fair value
is that fair values often incorporate value changes that are outside
of a manager’s control (Paul, 1992), so users might blame or give
credit to managers for uncontrollable events when those events
are highlighted in ?nancial statements through fair value measure-
ment. When managers are explicitly or implicitly incentivized by
measures that are uncontrollable, they demand a risk premium in
their compensation, leading to lower ?rm value.
ABHH abstract away from this issue in two ways. First, they
explicitly inform participants that managers have no control over
the exogenous demand shocks taking place throughout the ex-
periment. After learning about the shocks during training, partic-
ipants read that “management has no way of anticipating this eco-
nomic shock”. In each of the subsequent 20 rounds of the exper-
iment, participants again read that “the shock was unexpected”.
Second, participants allocate investments (and evaluate manager
quality) among two ?rms that are equally impacted by the exoge-
nous shock.
These design choices reduce the possibility that the fair value
condition, which highlights the exogenous shock more than does
the amortized cost condition, encourages participants to blame or
give credit to managers for the uncontrollable event. Rather, be-
cause the shock is exogenous, highlighting its salience helps partic-
ipants more easily identify whether managers respond to the shock
by investing in the division most favorably impacted by the shock.
Thus, in this setting, fair value measurement retains the bene?t
of helping investors see whether managers reinvest in divisions
experiencing a positive and persistent exogenous shock, but does
not retain the cost of encouraging investors to incorrectly attribute
blame or credit to the manager for experiencing the shock in the
?rst place. Future research could vary investors’ knowledge of the
extent to which shocks are exogenous, allowing investors to inap-
propriately blame or credit management for exogenous shocks that
are highlighted through fair value measurement.
2.3. Choice 3: Investors know that exogenous demand shocks are
persistent and actionable
Another concern with measuring assets and liabilities at fair
value is that investors may fail to appreciate the nature of differ-
ent economic shocks that are highlighted by fair values. In partic-
ular, fair values sometimes highlight economic shocks that man-
agers can act on to improve ?rm value (e.g., a ?ve-year demand
shock, as in ABHH) – “actionable shocks” – and other times high-
light economic shocks that managers cannot act on to improve
?rm value (e.g., discount rate changes) – “non-actionable shocks”.
In practice, it might not be clear whether shocks are actionable
or non-actionable. To the extent that investors misinterpret non-
actionable shocks as actionable, they may inappropriately punish
managers who do not act and reward those who do act. Con-
versely, to the extent that investors misinterpret actionable shocks
as non-actionable, they may inappropriately fail to punish man-
agers who do not act and fail to reward managers who do act.
ABHH abstract away from this issue by focusing on demand
shocks that persist for ?ve years and informing participants that
all shocks are actionable. In particular, ABHH inform participants
that an exogenous economic shock occurred during year 1 and that
participants should expect the shock to persist for ?ve years. In ad-
dition, participants read the following during experimental train-
ing: “ A good manager will use capital allocation decisions to
increase ?rm value in response to the shock. A bad manager
will use capital allocation decisions to decrease ?rm value in
response to the shock ” (emphasis as in original materials). After
each round of the experiment, participants receive feedback about
their decisions, reinforcing the notion that the exogenous shock
will persist and that ?rm value is enhanced when managers re-
spond to the shock by investing in the division experiencing a pos-
itive shock (or investing in the division not experiencing a negative
shock).
ABHH design their experiment in this way in order to eval-
uate participants’ stewardship assessments and investment deci-
sions against a normative benchmark. ABHH note that “because
the shock is expected to persist into the future, normatively the
relative quality of a manager depends on the extent to which the
manager allocates capital to maximize ?rm value in response to
the shock” (p. 16). However, these design choices likely encouraged
experimental participants to adopt the heuristic that they should
reward (punish) managers who invest in a division experiencing
a positive (negative) shock. In practice settings, the ambiguity of
the nature of shocks would make it more di?cult for investors to
unambiguously interpret investment in a division with a positive
shock as indicating good stewardship and investment in a division
with a negative shock as indicating poor stewardship. Future re-
search could vary the nature of economic shocks conveyed by fair
values to examine whether inappropriate assessments occur.
2.4. Choice 4: Investors know that fair value measures are free from
error and neutral
Opponents of recognizing assets and liabilities at fair value ar-
gue that fair value measures sometimes involve estimates that
are too uncertain or subjective to be easily veri?ed (e.g., see
Holthausen & Watts, 2001; Watts, 2003). These critics argue that
unveri?able estimates (or estimates that are too costly to verify)
are susceptible to manipulation by managers whose incentives de-
pend on those estimates, reducing the relevance of those measures
to investors.
ABHH abstract away from this issue by employing fair value
estimates that have no measurement uncertainty or bias. ABHH
inform participants: “As an investor, in order to maximize your
wealth, you should invest in the ?rm that will earn you the most
S.A. Emett, M.W. Nelson/ Accounting, Organizations and Society 46 (2015) 115–118 117
returns for that year.” In order to calculate returns (and thus par-
ticipants’ compensation), ABHH calculate a measure of ?rm value
that uses the same assumptions that underlie the fair value mea-
sures presented to experimental participants in the ?rms’ ?nan-
cial statements (see Appendix in ABHH). Accordingly, the fair value
measures in the experiment lack any measurement uncertainty or
bias – participants always earn higher payoffs by attending and re-
sponding to the information conveyed by the fair value measures.
In addition, participants receive feedback about their decisions af-
ter each of the 20 rounds, reinforcing the notion that they can earn
more money by relying on the fair value measures. Future research
could examine whether ABHH’s results generalize to a setting in
which the bene?t of timely gain and loss recognition is offset by
the cost of error or bias in fair value estimates.
2.5. Choice 5: Presentation
In ABHH’s experiment, investors are presented with simpli?ed
?nancial statements for two ?rms (see ABHH, Fig. 2), operational-
izing either amortized cost or fair value. The ?nancial statements
present a balance sheet in which the top line is gross assets and
the next line is either accumulated revaluation adjustments (in
the case of fair value) or accumulated depreciation (in the case
of amortized cost). Thus, the ?nancial statements present asset
reinvestments and fair value changes contiguously and saliently in
the fair value condition, helping investors detect whether reinvest-
ments and fair value changes covary positively or negatively. This
choice has two important implications.
First, the presentation used by ABHH does not appear to be
common practice. In a review of the ?nancial statements of sev-
eral IFRS ?rms reporting revaluation adjustments, we were unable
to ?nd a ?rm reporting revaluation adjustments on the face of the
balance sheet. All ?rms instead opted to report only net assets on
the face of the balance sheet and then disaggregate assets in the
notes to the ?nancial statements. This suggests that ABHH’s experi-
mental results could inform discussions of ?nancial statement pre-
sentation, should standard setters wish to consider the effects of
alternative presentations that render revaluation adjustment more
salient, but that those results don’t map directly into the presenta-
tion formats observed in current practice.
Second, we believe that the presentation used by ABHH sug-
gests an interesting alternative theoretical explanation for ABHH’s
results. To elaborate, we start by brie?y summarizing ABHH’s ex-
planation, and then describe the alternative.
At a broad theoretical level, ABHH argue that fair value mea-
sures are inherently more transparent than are standardized cost
measures (p. 4). Later in the paper, ABHH provide a more spe-
ci?c explanation of the process by which fair value measures more
transparently provide information to investors. First, ABHH de-
scribe two advantages of fair value measures relative to amortized
cost measures: (1) fair value measures provide more timely infor-
mation, and (2) fair value measures are more comparable across
?rms. Second, ABHH describe how these two advantages shape the
quality of investors’ stewardship assessments. They argue that in-
vestors compare fair values across ?rms to observe whether fair
value changes are correlated or uncorrelated across ?rms. Investors
then use these correlations to identify and ignore value changes
driven by exogenous forces so they can focus on those driven by
managerial actions.
The investor judgment process that ABHH describe likely would
be challenging for participants in ABHH’s experimental setting.
Looking at the experimental stimuli (see Fig. 2 of ABHH), partic-
ipants can see that both ?rms report a revaluation adjustment
of comparable magnitude. The revaluation adjustment aggregates
value changes driven by exogenous shocks with those driven by
managerial actions. Hence, participants cannot easily isolate value
changes driven by manager actions simply by observing correla-
tions in the revaluation adjustment across ?rms.
An alternative explanation for ABHH’s results focuses not on de-
tecting between-?rm covariation of shocks for purposes of ignor-
ing shocks, but rather focuses on detecting covariation of shocks
and reinvestment within a particular ?rm. Participants’ task was
to identify which of the two ?rms invested more in the division
that experienced a positive shock (or invested more in the division
that did not experience a negative shock). In the fair value con-
dition, asset reinvestments and shocks are presented on the ?rst
two lines of the ?nancial statements, but in the amortized cost
condition, these elements are presented in different ?nancial state-
ments. Thus, within a particular ?rm, ABHH’s fair value condition
helps participants to detect whether managers reinvested in divi-
sions experiencing a positive shock (or reinvested in divisions not
experiencing a negative shock).
Future research could distinguish our alternative explanation
from ABHH’s explanation. First, ABHH’s experimental materials
could be modi?ed so that participants view the ?nancial state-
ments of a single ?rm rather than two ?rms. Such a design pre-
cludes ABHH’s theoretical explanation (which relies on investors
observing correlations across ?rms), but does not preclude ours.
Second, a study could hold constant the proximity of information
about economic shocks to information about managers’ reinvest-
ments. Such a design precludes our theoretical explanation (which
relies on proximity differing across conditions), but does not pre-
clude theirs.
3. Conclusion
ABHH provide evidence that fair value information can be bene-
?cial for stewardship assessments, and call for research that inves-
tigates “not whether, but under what conditions, amortized-cost-
based ?nancial statements or fair-value-based ?nancial statements
are likely to provide more useful information” (p. 31). We believe
that ABHH make an important contribution by challenging an im-
portant premise underlying a longstanding debate in the literature.
We echo their call for future research and outline ?ve methodolog-
ical choices made by ABHH that likely in?uence the relative use-
fulness of fair value and amortized cost information. With those
design choices in mind, we discuss an alternative theoretical ex-
planation for ABHH’s results and suggest future research that can
differentiate between these alternative explanations. We look for-
ward to continued research in this area that sheds light on the rel-
ative merits of alternative measurement approaches for achieving
alternative ?nancial reporting objectives.
References
Anderson, S., Brown, J. L., Hodder, L., & Hopkins, P. E. (in press). The effect of
alternative accounting measurement bases on investors’ assessments of man-
agement stewardship. Accounting, Organizations and Society,http://dx.doi.org/10.
1016/j.aos.2015.03.007.
Bushman, R., Engel, E., & Smith, A. (2006). An analysis of the relation between
the stewardship and valuation roles of earnings. Journal of Accounting Research,
44(1), 53–83.
Bushman, R. M., & Indjejikian, R. J. (1993). Stewardship value of ‘distorted’ account-
ing disclosures. The Accounting Review, 68(4), 765–782.
Chen, R. S. (1975). Social and ?nancial stewardship. The Accounting Review, 50(3),
533–543.
Cormier, D., Magnan, M., & Morard, B. (2000). The contractual and value relevance
of reported earnings in a dividend-focused environment. The European Account-
ing Review, 9(3), 387–417.
Gjesdal, F. G. (1981). Accounting for stewardship. Journal of Accounting Research,
19(1), 208–231.
Hodder, L., Hopkins, P. E., & Schipper, K. (in press). Fair value measurement in ?-
nancial reporting. Foundations and Trends in Accounting.
118 S.A. Emett, M.W. Nelson/ Accounting, Organizations and Society 46 (2015) 115–118
Holthausen, R. W., & Watts, R. L. (2001). The relevance of the value-relevance lit-
erature for ?nancial accounting standard setting. Journal of Accounting and Eco-
nomics, 31, 3–75.
O’Connell, V. (2007). Re?ections on stewardship reporting. Accounting Horizons,
21(2), 215–227.
Paul, J. M. (1992). On the e?ciency of stock-based compensation. The Review of Fi-
nancial Studies, 5(3), 471–502.
Watts, R. L. (2003). Conservatism in accounting part 1: Explanations and implica-
tions. Accounting Horizons, 17(3), 207–221.
doc_582057460.pdf
Anderson,Brown,Hodder,andHopkins(ABHH,inpress)provideexperimentalevidencethatinvestorsarebetterabletoassessmanagementstewardshipandfirmvaluewhenfinancialstatementshighlightfir-valueinformationratherthanamortized-costinformation.Wefocusourdiscussiononfivemethod-ologicalchoicesthattheauthorsmake.WeconsiderhowthosechoicesaffectthegeneralityofABHH’sresults,consideranalternativetheoreticalexplanationforthoseresults,andsuggestopportunitiesforfutureresearchthatbuildonABHH’swork.
Accounting, Organizations and Society 46 (2015) 115–118
Contents lists available at ScienceDirect
Accounting, Organizations and Society
journal homepage: www.elsevier.com/locate/aos
Discussion of “The effect of alternative accounting measurement bases
on investors’ assessments of managers’ stewardship”
R
Scott A. Emett, Mark W. Nelson
?
S.C. Johnson Graduate School of Management, Cornell University, United States
a r t i c l e i n f o
Article history:
Available online 29 August 2015
a b s t r a c t
Anderson, Brown, Hodder, and Hopkins (ABHH, in press) provide experimental evidence that investors
are better able to assess management stewardship and ?rm value when ?nancial statements highlight
fair-value information rather than amortized-cost information. We focus our discussion on ?ve method-
ological choices that the authors make. We consider how those choices affect the generality of ABHH’s
results, consider an alternative theoretical explanation for those results, and suggest opportunities for
future research that build on ABHH’s work.
© 2015 Elsevier Ltd. All rights reserved.
1. Introduction
Anderson, Brown, Hodder, and Hopkins (hereafter ABHH, in
press) seek to inform a longstanding debate about the extent to
which fair value and amortized cost measurement approaches aid
?nancial statement users in assessing stewardship and in mak-
ing investment decisions. As argued by ABHH and others (e.g., see
Hodder, Hopkins, & Schipper, in press), the accounting literature
often assumes (but does not offer much evidence) that histori-
cal amortized-cost-based ?nancial information is inherently supe-
rior for stewardship purposes relative to fair-value-based ?nancial
information. ABHH provide evidence that fair value information
yields better stewardship assessments than does amortized cost in-
formation in their experimental setting, informing this debate.
One way to view ABHH’s contribution is that they invalidate
the assumption that amortized-cost information is always superior
to fair value information for assessing stewardship by demonstrat-
ing an instance in which the converse is true. Their work should
naturally lead to further research considering when each measure-
ment attribute (amortized cost or fair value) is more helpful for
assessing stewardship and/or providing decision-useful information
to investors. Therefore, we focus our discussion on ?ve of ABHH’s
methodological choices that likely affect the generality of their re-
R
We thank the authors of Anderson et al. (2015) (Spencer Anderson, Jason Brown,
Leslie Hodder, and Pat Hopkins), as well as Rob Bloom?eld and Bob Libby, for pro-
viding comments on this discussion. We gratefully acknowledge Deloitte & Touche
LLP’s ?nancial support of the AOS Estimates Conference.
?
Corresponding author.
E-mail address: [email protected] (M.W. Nelson).
sults and that offer opportunities for future research that modi?es
these choices. We also discuss an alternative theoretical explana-
tion for ABHH’s results which future research could seek to differ-
entiate from the explanation that ABHH provide.
2. Methodological choices and generality of results
2.1. Choice 1: De?ning good stewardship as actions that maximize
the present value of future cash ?ows
De?ning stewardship requires specifying the rights and respon-
sibilities of an agent in relation to a principal (Bushman, Engel, &
Smith, 2006; Bushman & Indjejikian, 1993; Chen, 1975; Cormier,
Magnan, & Morard, 2000; Gjesdal, 1981; O’Connell, 2007). ABHH
create a setting in which participants assume the role of investor
and receive higher payoffs by investing more in (or assessing
higher manager quality of) ?rms that increase the most in ?rm
value. In this setting, managers have clear rights – the right to
use the investor’s capital – and clear responsibilities – the respon-
sibility to increase ?rm value. ABHH’s de?nition of stewardship
matches this setting: they de?ne stewardship as “the actions of
managers in the discharge of their responsibilities to preserve the
value of investors’ capital investment and to earn a commensurate
return on their investment” (p. 1–2). This de?nition links stew-
ardship directly to ?rm value: decisions that increase ?rm value
unequivocally imply good stewardship, and decisions that decrease
?rm value unequivocally imply poor stewardship. A measurement
basis that is useful to investors in assessing stewardship in this
setting should always similarly be useful in making investment al-http://dx.doi.org/10.1016/j.aos.2015.08.003
0361-3682/© 2015 Elsevier Ltd. All rights reserved.
116 S.A. Emett, M.W. Nelson/ Accounting, Organizations and Society 46 (2015) 115–118
location decisions. Thus, ABHH’s study does not inform us about
how stewardship assessments might differ from ?rm valuation
assessments.
Different settings might operationalize different notions of
stewardship and produce conclusions from a stewardship perspec-
tive that differ from those of a ?rm valuation perspective. For ex-
ample, Watts (2003) argues that ?nancial accounting has evolved
to be conservative because many parties that contract with the
?rm have asymmetric payoffs from their contracts. Those parties
are concerned about the ?rm’s ability to reach some threshold
level of pro?tability, but are less concerned about the extent to
which the ?rm exceeds that threshold of pro?tability. Such parties
might view a manager as being a better steward if the manager
invests in lower-risk projects after meeting a threshold of prof-
itability, even if that means turning down high risk projects with
a positive net present value that would enhance ?rm value. Sim-
ilarly, investors for whom company-speci?c risks are undiversi?-
able might prefer manager behaviors that are more conservative
than might otherwise be preferred. These examples imply a de?ni-
tion of stewardship that emphasizes capital preservation, such that
some decisions that increase expected ?rm value may imply poor
stewardship, and some decisions that decrease expected ?rm value
may imply good stewardship. Future research could examine these
settings in order to operationalize other notions of stewardship.
2.2. Choice 2: Investors know that demand shocks are completely
exogenous
One concern with measuring assets and liabilities at fair value
is that fair values often incorporate value changes that are outside
of a manager’s control (Paul, 1992), so users might blame or give
credit to managers for uncontrollable events when those events
are highlighted in ?nancial statements through fair value measure-
ment. When managers are explicitly or implicitly incentivized by
measures that are uncontrollable, they demand a risk premium in
their compensation, leading to lower ?rm value.
ABHH abstract away from this issue in two ways. First, they
explicitly inform participants that managers have no control over
the exogenous demand shocks taking place throughout the ex-
periment. After learning about the shocks during training, partic-
ipants read that “management has no way of anticipating this eco-
nomic shock”. In each of the subsequent 20 rounds of the exper-
iment, participants again read that “the shock was unexpected”.
Second, participants allocate investments (and evaluate manager
quality) among two ?rms that are equally impacted by the exoge-
nous shock.
These design choices reduce the possibility that the fair value
condition, which highlights the exogenous shock more than does
the amortized cost condition, encourages participants to blame or
give credit to managers for the uncontrollable event. Rather, be-
cause the shock is exogenous, highlighting its salience helps partic-
ipants more easily identify whether managers respond to the shock
by investing in the division most favorably impacted by the shock.
Thus, in this setting, fair value measurement retains the bene?t
of helping investors see whether managers reinvest in divisions
experiencing a positive and persistent exogenous shock, but does
not retain the cost of encouraging investors to incorrectly attribute
blame or credit to the manager for experiencing the shock in the
?rst place. Future research could vary investors’ knowledge of the
extent to which shocks are exogenous, allowing investors to inap-
propriately blame or credit management for exogenous shocks that
are highlighted through fair value measurement.
2.3. Choice 3: Investors know that exogenous demand shocks are
persistent and actionable
Another concern with measuring assets and liabilities at fair
value is that investors may fail to appreciate the nature of differ-
ent economic shocks that are highlighted by fair values. In partic-
ular, fair values sometimes highlight economic shocks that man-
agers can act on to improve ?rm value (e.g., a ?ve-year demand
shock, as in ABHH) – “actionable shocks” – and other times high-
light economic shocks that managers cannot act on to improve
?rm value (e.g., discount rate changes) – “non-actionable shocks”.
In practice, it might not be clear whether shocks are actionable
or non-actionable. To the extent that investors misinterpret non-
actionable shocks as actionable, they may inappropriately punish
managers who do not act and reward those who do act. Con-
versely, to the extent that investors misinterpret actionable shocks
as non-actionable, they may inappropriately fail to punish man-
agers who do not act and fail to reward managers who do act.
ABHH abstract away from this issue by focusing on demand
shocks that persist for ?ve years and informing participants that
all shocks are actionable. In particular, ABHH inform participants
that an exogenous economic shock occurred during year 1 and that
participants should expect the shock to persist for ?ve years. In ad-
dition, participants read the following during experimental train-
ing: “ A good manager will use capital allocation decisions to
increase ?rm value in response to the shock. A bad manager
will use capital allocation decisions to decrease ?rm value in
response to the shock ” (emphasis as in original materials). After
each round of the experiment, participants receive feedback about
their decisions, reinforcing the notion that the exogenous shock
will persist and that ?rm value is enhanced when managers re-
spond to the shock by investing in the division experiencing a pos-
itive shock (or investing in the division not experiencing a negative
shock).
ABHH design their experiment in this way in order to eval-
uate participants’ stewardship assessments and investment deci-
sions against a normative benchmark. ABHH note that “because
the shock is expected to persist into the future, normatively the
relative quality of a manager depends on the extent to which the
manager allocates capital to maximize ?rm value in response to
the shock” (p. 16). However, these design choices likely encouraged
experimental participants to adopt the heuristic that they should
reward (punish) managers who invest in a division experiencing
a positive (negative) shock. In practice settings, the ambiguity of
the nature of shocks would make it more di?cult for investors to
unambiguously interpret investment in a division with a positive
shock as indicating good stewardship and investment in a division
with a negative shock as indicating poor stewardship. Future re-
search could vary the nature of economic shocks conveyed by fair
values to examine whether inappropriate assessments occur.
2.4. Choice 4: Investors know that fair value measures are free from
error and neutral
Opponents of recognizing assets and liabilities at fair value ar-
gue that fair value measures sometimes involve estimates that
are too uncertain or subjective to be easily veri?ed (e.g., see
Holthausen & Watts, 2001; Watts, 2003). These critics argue that
unveri?able estimates (or estimates that are too costly to verify)
are susceptible to manipulation by managers whose incentives de-
pend on those estimates, reducing the relevance of those measures
to investors.
ABHH abstract away from this issue by employing fair value
estimates that have no measurement uncertainty or bias. ABHH
inform participants: “As an investor, in order to maximize your
wealth, you should invest in the ?rm that will earn you the most
S.A. Emett, M.W. Nelson/ Accounting, Organizations and Society 46 (2015) 115–118 117
returns for that year.” In order to calculate returns (and thus par-
ticipants’ compensation), ABHH calculate a measure of ?rm value
that uses the same assumptions that underlie the fair value mea-
sures presented to experimental participants in the ?rms’ ?nan-
cial statements (see Appendix in ABHH). Accordingly, the fair value
measures in the experiment lack any measurement uncertainty or
bias – participants always earn higher payoffs by attending and re-
sponding to the information conveyed by the fair value measures.
In addition, participants receive feedback about their decisions af-
ter each of the 20 rounds, reinforcing the notion that they can earn
more money by relying on the fair value measures. Future research
could examine whether ABHH’s results generalize to a setting in
which the bene?t of timely gain and loss recognition is offset by
the cost of error or bias in fair value estimates.
2.5. Choice 5: Presentation
In ABHH’s experiment, investors are presented with simpli?ed
?nancial statements for two ?rms (see ABHH, Fig. 2), operational-
izing either amortized cost or fair value. The ?nancial statements
present a balance sheet in which the top line is gross assets and
the next line is either accumulated revaluation adjustments (in
the case of fair value) or accumulated depreciation (in the case
of amortized cost). Thus, the ?nancial statements present asset
reinvestments and fair value changes contiguously and saliently in
the fair value condition, helping investors detect whether reinvest-
ments and fair value changes covary positively or negatively. This
choice has two important implications.
First, the presentation used by ABHH does not appear to be
common practice. In a review of the ?nancial statements of sev-
eral IFRS ?rms reporting revaluation adjustments, we were unable
to ?nd a ?rm reporting revaluation adjustments on the face of the
balance sheet. All ?rms instead opted to report only net assets on
the face of the balance sheet and then disaggregate assets in the
notes to the ?nancial statements. This suggests that ABHH’s experi-
mental results could inform discussions of ?nancial statement pre-
sentation, should standard setters wish to consider the effects of
alternative presentations that render revaluation adjustment more
salient, but that those results don’t map directly into the presenta-
tion formats observed in current practice.
Second, we believe that the presentation used by ABHH sug-
gests an interesting alternative theoretical explanation for ABHH’s
results. To elaborate, we start by brie?y summarizing ABHH’s ex-
planation, and then describe the alternative.
At a broad theoretical level, ABHH argue that fair value mea-
sures are inherently more transparent than are standardized cost
measures (p. 4). Later in the paper, ABHH provide a more spe-
ci?c explanation of the process by which fair value measures more
transparently provide information to investors. First, ABHH de-
scribe two advantages of fair value measures relative to amortized
cost measures: (1) fair value measures provide more timely infor-
mation, and (2) fair value measures are more comparable across
?rms. Second, ABHH describe how these two advantages shape the
quality of investors’ stewardship assessments. They argue that in-
vestors compare fair values across ?rms to observe whether fair
value changes are correlated or uncorrelated across ?rms. Investors
then use these correlations to identify and ignore value changes
driven by exogenous forces so they can focus on those driven by
managerial actions.
The investor judgment process that ABHH describe likely would
be challenging for participants in ABHH’s experimental setting.
Looking at the experimental stimuli (see Fig. 2 of ABHH), partic-
ipants can see that both ?rms report a revaluation adjustment
of comparable magnitude. The revaluation adjustment aggregates
value changes driven by exogenous shocks with those driven by
managerial actions. Hence, participants cannot easily isolate value
changes driven by manager actions simply by observing correla-
tions in the revaluation adjustment across ?rms.
An alternative explanation for ABHH’s results focuses not on de-
tecting between-?rm covariation of shocks for purposes of ignor-
ing shocks, but rather focuses on detecting covariation of shocks
and reinvestment within a particular ?rm. Participants’ task was
to identify which of the two ?rms invested more in the division
that experienced a positive shock (or invested more in the division
that did not experience a negative shock). In the fair value con-
dition, asset reinvestments and shocks are presented on the ?rst
two lines of the ?nancial statements, but in the amortized cost
condition, these elements are presented in different ?nancial state-
ments. Thus, within a particular ?rm, ABHH’s fair value condition
helps participants to detect whether managers reinvested in divi-
sions experiencing a positive shock (or reinvested in divisions not
experiencing a negative shock).
Future research could distinguish our alternative explanation
from ABHH’s explanation. First, ABHH’s experimental materials
could be modi?ed so that participants view the ?nancial state-
ments of a single ?rm rather than two ?rms. Such a design pre-
cludes ABHH’s theoretical explanation (which relies on investors
observing correlations across ?rms), but does not preclude ours.
Second, a study could hold constant the proximity of information
about economic shocks to information about managers’ reinvest-
ments. Such a design precludes our theoretical explanation (which
relies on proximity differing across conditions), but does not pre-
clude theirs.
3. Conclusion
ABHH provide evidence that fair value information can be bene-
?cial for stewardship assessments, and call for research that inves-
tigates “not whether, but under what conditions, amortized-cost-
based ?nancial statements or fair-value-based ?nancial statements
are likely to provide more useful information” (p. 31). We believe
that ABHH make an important contribution by challenging an im-
portant premise underlying a longstanding debate in the literature.
We echo their call for future research and outline ?ve methodolog-
ical choices made by ABHH that likely in?uence the relative use-
fulness of fair value and amortized cost information. With those
design choices in mind, we discuss an alternative theoretical ex-
planation for ABHH’s results and suggest future research that can
differentiate between these alternative explanations. We look for-
ward to continued research in this area that sheds light on the rel-
ative merits of alternative measurement approaches for achieving
alternative ?nancial reporting objectives.
References
Anderson, S., Brown, J. L., Hodder, L., & Hopkins, P. E. (in press). The effect of
alternative accounting measurement bases on investors’ assessments of man-
agement stewardship. Accounting, Organizations and Society,http://dx.doi.org/10.
1016/j.aos.2015.03.007.
Bushman, R., Engel, E., & Smith, A. (2006). An analysis of the relation between
the stewardship and valuation roles of earnings. Journal of Accounting Research,
44(1), 53–83.
Bushman, R. M., & Indjejikian, R. J. (1993). Stewardship value of ‘distorted’ account-
ing disclosures. The Accounting Review, 68(4), 765–782.
Chen, R. S. (1975). Social and ?nancial stewardship. The Accounting Review, 50(3),
533–543.
Cormier, D., Magnan, M., & Morard, B. (2000). The contractual and value relevance
of reported earnings in a dividend-focused environment. The European Account-
ing Review, 9(3), 387–417.
Gjesdal, F. G. (1981). Accounting for stewardship. Journal of Accounting Research,
19(1), 208–231.
Hodder, L., Hopkins, P. E., & Schipper, K. (in press). Fair value measurement in ?-
nancial reporting. Foundations and Trends in Accounting.
118 S.A. Emett, M.W. Nelson/ Accounting, Organizations and Society 46 (2015) 115–118
Holthausen, R. W., & Watts, R. L. (2001). The relevance of the value-relevance lit-
erature for ?nancial accounting standard setting. Journal of Accounting and Eco-
nomics, 31, 3–75.
O’Connell, V. (2007). Re?ections on stewardship reporting. Accounting Horizons,
21(2), 215–227.
Paul, J. M. (1992). On the e?ciency of stock-based compensation. The Review of Fi-
nancial Studies, 5(3), 471–502.
Watts, R. L. (2003). Conservatism in accounting part 1: Explanations and implica-
tions. Accounting Horizons, 17(3), 207–221.
doc_582057460.pdf