Accounting Conservatism: A Literature Review: Yuxiang Zhong, Wanli Li

Accounting Conservatism: A Literature Review
Yuxiang Zhong, School of Management, Xi’an Jiaotong University

Wanli Li, School of Management, Xi’an Jiaotong University and Accounting School, Shanghai
University of International Business and Economics
Accounting conservatism is one of the most important properties of financial reporting. The goal of this article is
to gain a better understanding of accounting conservatism. We explicate the evolution of conservatism over its
long history. Accounting conservatism is indispensable because the main parties of a firm demand conservatism
to mitigate agency costs. Various methods are used to measure accounting conservatism, which include balance
sheet measures, income statement measures and earnings/stock return relation measures. Empirical research
into accounting conservatism has flourished over the last two decades and we focus on the cross-sectional and
time-series variations in conservatism. We conclude that accounting conservatism is important and cannot be
excluded from accounting standards.
ccounting conservatism is one of the most ‘good news’ and ‘bad news’, where ‘good news’ indicates
A important properties of financial reporting

(Sterling 1970). Accounting standards usually
include conservatism principles, such as lower of cost
positive expected cash flows and ‘bad news’ indicates
negative expected cash flows. Basu (1997) interprets ac-
counting conservatism to capture accountants’ tendency
or market accounting for inventory and impairment ac- to require a higher degree of verification for recognising
counting for tangible and intangible assets (Ball et al. good news than bad news in financial statements. There-
2000a; Wyatt et al. 2001; Kabir and Laswad 2014). Over fore, ‘good news’ is recognised gradually over time, while
the last two decades, research on conservatism has flour- ‘bad news’ is incorporated into earnings in a timely man-
ished, and the findings show that financial reporting ner in the current period. Both ‘good news’ and ‘bad
is conservative in many markets (Basu 1997; Ball et al. news’ are usually recognised through accruals.
2000a, 2000b, 2003). Nevertheless, conservatism (pru- In this paper, we aim to review research findings re-
dence) is excluded from the joint conceptual framework lated to the following questions.
proposed by the International Accounting Standards
Board (IASB) and the Financial Accounting Standards 1. What is accounting conservatism, and how do we
Board (FASB) in 2010, even though it plays an important measure it (Part I of Figure 1)?
role in financial reporting. In this paper, we summarise 2. Which factors affect accounting conservatism, and
research findings on accounting conservatism to high- what are the explanations for accounting conser-
light its importance and promote future research on the vatism (Part II of Figure 1)?
topic. 3. How does accounting conservatism affect investment
Accounting systems are used to reflect firms’ economic efficiency and capital costs (Part III of Figure 1)?
circumstances, and one essential issue of accounting con-
cerns the means by which economic values are mapped to
earnings (see Figure 1). We decompose a firm’s economic The Evolution of Conservatism
value into a realised and unrealised part. The realised part
is objective and is recognised in earnings through cash Accounting conservatism has a long history. In this
flows, whereas the unrealised part must be predicted and section, we outline the definitions and classifica-
subjective. The accounting process occurs on an accrual tions of accounting conservatism. We then discuss the
basis, which plays a central role in financial reporting.
How the unrealised part is recognised in earnings deter-
mines the properties of financial reporting, such as its
Correspondence: Yuxiang Zhong, School of Management,
relevance and reliability. Accounting conservatism is an Xi’an Jiaotong University, No. 28, West Xianning Road,
important principle in the accounting process. The unre- Xi’an, Shaanxi, P.R. China. Tel.: +86-15686002716; email:
alised part of economic value can be further divided into xjtuzhongyuxiang@outlook.com
Australian Accounting Review No. 0 Vol. 0 Issue 0 2016 doi: 10.1111/auar.12107 1

Accounting Conservatism: A Literature Review Y. Zhong & W. Li
Figure 1 The conceptual framework of the review
relation between accounting conservatism and earnings reach contradicting conclusions. Basu (1997) proposes a
management. new definition of accounting conservatism and designs
a simple but insightful model to measure it. Both the
definition and the model are widely used. He interprets
Definitions of conservatism
conservatism to capture accountants’ tendency to re-
quire a higher degree of verification for recognising good
Both scholars and accounting standard setters have news compared with bad news in financial statements.
attempted to provide a generally accepted definition of Earnings reflect bad news more quickly than good news
accounting conservatism. Bliss (1924) defines conser- under conservatism principles. This definition accounts
vatism as ‘anticipat[ing] no profits but anticipat[ing] for different types of economic income and emphasises
all losses’. This definition is constructive but too simple. the timeliness of recognising economic losses, which is
Devine (1963) links accounting conservatism to ac- called conditional conservatism. Ball et al. (2000a) point
counting information users and firms’ goals. He defines out that this definition actually defines conservatism as
accounting conservatism as a rule leading to lower aver- the extent to which current-period accounting income
age expectations of goal fulfilment in comparison with asymmetrically incorporates economic losses, relative to
alternative measuring and reporting rules. Although economic gains.
this definition is more formal, uncertainty remains
regarding firms’ goals and the number of accounting
information users. Regulators also devote attention to Classifications of conservatism
accounting conservatism. Statement of concepts No. 2
by the FASB (1975) defines conservatism as ‘[a] prudent Scholars have begun distinguishing different types of
reaction to uncertainty ... If two estimates of amounts to accounting conservatism since Basu (1997) in order to
be received or paid in the future are about equally likely, conduct more targeted research. Confusion between the
conservatism dictates using the less optimistic estimate’. unconditional and conditional versions of conservatism
Watts and Zimmerman (1986) refer to conservatism as helps explain why conservatism is a controversial aspect
the rule that accountants should report the lowest value of accounting, despite its long-standing influence on ac-
among the possible alternative values for assets and the counting practice (Ball and Shivakumar 2005). There-
highest alternative value for liabilities. Further, revenue fore, it is very important to distinguish different types
should be recognised later than sooner, and expenses, of accounting conservatism and to determine how they
sooner than later. This definition accounts for financial relate to each other.
reporting and suggests that conservatism is a feature of A number of articles provide classifications of
financial reporting. conservatism (Basu 1997; Beekes et al. 2004; Chandra
Although these definitions are useful, they do not 2011). Beaver and Ryan (2005) classify accounting
consider the actual economic environment and do not conservatism into unconditional conservatism and
distinguish different types of conservatism. As a result, conditional conservatism and elaborate the relation and
measuring conservatism is difficult, and scholars thus distinction between them. Unconditional conservatism,
2 Australian Accounting Review

C 2016 CPA Australia
Y. Zhong & W. Li Accounting Conservatism: A Literature Review
also called news-independent or ex ante conservatism, discretion in conservatism. They document that non-
means that aspects of the accounting process deter- discretionary conservatism – or, more accurately, ‘nor-
mined at the inception of assets and liabilities yield the mal conservatism’ – is a level of conservatism that is
expected unrecorded goodwill. Examples of uncondi- determined by the firm’s economic circumstances, given
tional conservatism include the immediate expensing the discretion in accounting principles. The level of di-
of the cost of internally generated intangible assets and vergence from the normal level then is defined as discre-
the amortisation of long-lived assets at a rate above tionary conservatism. This classification suggests that
the expected economic amortisation rate. Conditional there is an ideal level of conservatism, determined by
conservatism, also called news-dependent or ex post the firm’s economic circumstances and implemented by
conservatism, means that book values are recorded managers. The actual level of conservatism always di-
under sufficiently adverse circumstances but are not verges from the normal level of conservatism because
recorded under favourable circumstances. Examples managers’ interests are not in line with the firm’s.
of conditional conservatism include lower of cost
or market accounting for inventory and impairment
accounting for long-lived tangible and intangible assets Accounting conservatism and earnings management
(Ji 2013). Both of these types of conservatism lead to
the book value of net assets being understated relative Hanna (2002) argues that earnings management can ex-
to their market value (Kabir and Laswad 2014). plain Basu’s (1997) empirical findings. Earnings man-
In contrast to unconditional conservatism, condi- agement constitutes a manager’s choices regarding ac-
tional conservatism carries new information (Ball et al. counting policies, or real actions that affect earnings so
2013a) and depends on the economic environment faced as to achieve specific reported earnings objectives (Scott
by firms. Usually, conditional conservatism requires 2012). Based on this definition, Scott (2012) lists four
economic losses to be recognised in a more timely common manifestations of earnings management: ‘big
manner than economic gains. Using a mathematical bath’, income minimisation, income maximisation and
model, Beaver and Ryan (2005) show that unconditional income smoothing. Earnings management can also lead
conservatism is a primary source of unrecorded good- to an understatement of assets, if assets are inappropri-
will, which constitutes a form of ‘accounting slack’ that ately recorded and liabilities and transitory losses are
pre-empts the application of conditional conservatism increased. Nevertheless, earnings management and ac-
unless news is sufficiently bad to use up the slack. counting conservatism significantly differ.
Another salient difference between unconditional and Basu (1995) points out that conservatism dictates the
conditional conservatism concerns their impact on relevant and accepted set of accounting principles used
contracting efficiency. Ball and Shivakumar (2005) in accounting practice. Earnings management then de-
point out that conditional conservatism can improve termines how managers use these accounting principles
contracting and investment efficiency through the in financial reporting. For example, ASC 350 (FASB,
timely recognition of losses, which would restrict man- 2009) requires an impairment test for goodwill and
agers’ opportunistic actions. However, unconditional indefinite-lived intangibles when their fair values are
conservatism would prevent such conditional conser- more likely than not to be materially less than their car-
vatism from improving contracting efficiency and could rying amounts. However, managers usually determine
even distort financial reporting used by investors. Qiang the calculation of fair value.
(2007) provides empirical evidence that conditional With conservatism, net assets values are persistently
conservatism and unconditional conservatism have a understated under asymmetric treatment of gains and
negative relation with each other and play different roles losses (Watts 2003a). Watts (2003a) claims that ‘conser-
in firms. Researchers therefore must trade-off between vatism refers to the cumulative financial effects repre-
conditional and unconditional conservatism. sented in the balance sheet and to income or earnings
From another perspective, Lawrence et al. (2013) dis- cumulated since the firm began operation’. By contrast,
tinguish between discretionary and non-discretionary accrual-based earnings management must follow the ac-
conservatism in a classification of conservatism anal- crual reversal method, which means that upwards earn-
ogous to the case of discretionary earnings and ings management in the current period will lead to the
non-discretionary earnings. They argue that non- same amount of downwards earnings management in
discretionary conservatism results from the unbiased future periods and vice versa (Scott 2012). Hence, in
application of accounting principles and define discre- contrast to conservatism, earnings management is un-
tionary conservatism as conservatism arising from pur- likely to persistently understate net asset values.
poseful intervention in the financial reporting process Conservatism decreases managers’ incentives to in-
to adjust the amount and timing of conservatism ac- flate reported earnings for higher compensation and to
counting. Roychowdhury and Martin (2013) agree with invest in negative net present value (hereafter NPV)
this classification but take an alternative viewpoint on projects. In addition, conservatism triggers contract

C 2016 CPA Australia Australian Accounting Review 3
covenants in time and requires managers to stop sons. Accounting information is produced by firms and
poor performance projects ex post. Conservatism thus audited by auditors, so it is relatively reliable. Such in-
restricts managers’ actions in overstating earnings and formation is also easy for lenders to obtain and costless.
pursuing their own interests. However, because man- Moreover, earnings or net assets are efficient measures
agers’ compensation is usually tied to earnings, managers of firm or manager performance (Scott 2012). Empirical
will always overstate earnings if possible. ‘Big bath’ is a evidence shows that financial covenants are included in
rare phenomenon, and it generally does not occur dur- both bank-loan and public-debt contracting in different
ing short periods. Thus, earnings management seems to markets (Mather 1999). For example, covenants often in-
be more likely to lead to overstating than understating clude thresholds that restrict the payment of dividends
net assets. and thresholds in terms of the fixed-charge coverage ra-
Several papers investigate the relation between tio, net worth and leverage (Nikolaev 2010).
conservatism and earnings management by building a The debt contracting protects debtholders’ interests
theoretical model; however, the conclusions drawn are by transferring control rights to them when covenants
inconsistent. Chen et al. (2007) and Gao (2013) argue are violated. Only when the accounting information
that accounting conservatism improves contracting used by contracts is timely can debt contracting suffi-
efficiency and curbs managers’ incentives to inflate ciently protect debtholders’ interests. Debtholders pre-
earnings. However, Jackson and Liu (2010) investigate fer to recognise losses rather than gains in a timely
the allowance for uncollectible accounts and find that ac- manner because of asymmetric payoffs. They do not
counting conservatism is also likely to facilitate earnings care much about timely gain recognition because they
management. Although earnings management is similar cannot acquire additional payoffs apart from the prin-
to conservatism and can partly explain the evidence sup- cipal and interest. However, once a firm suffers from
ported by conservatism, it cannot replace conservatism. financial trouble, debtholders will lose their wealth.
Conservatism and earnings management are neither en- Therefore, they demand timely recognition of losses.
tirely mutually exclusive nor entirely one phenomenon. Conservative accounting has evolved to improve the ef-
ficiency of debt contracting by triggering covenant vio-
lations via timely loss recognition and by requiring high
Demand for Conservatism verifiability for gains (Watts 2003a). With timely loss
recognition, losses will be reflected in financial report-
Financial reporting is used for both valuation by equity ing in a timely manner, and covenant violations will
investors and performance evaluation, referred to as effi- be more likely to occur. Debtholders can then take ac-
cient contracting. Conservative financial reporting is not tion to protect their wealth in a timely manner. Further-
suitable for valuation because of the potential of provid- more, when gains are recognised with high verifiabil-
ing biased information to investors, while conservatism ity, it is difficult for managers to inflate earnings or net
is necessary for efficient contracting (Kothari et al. 2010). assets. Consequently, debtholders demand accounting
Watts (2003a) proposes four explanations for conser- conservatism.
vatism: a contracting explanation, a litigation explana- Debtholders’ demand for conservatism is supported
tion, an income tax explanation and a regulatory expla- by numerous empirical studies. Ahmed et al. (2002)
nation. From another perspective, we describe a different show that the financial reporting of firms with severe
party’s demand for conservatism in this paper. We argue conflicts over dividend policy is more conservative
that four main parties demand conservatism: debthold- and that the cost of debt is lower in firms with more
ers, shareholders, auditors and regulators. Managers conservative financial reporting. Zhang (2008) provides
play a special role in accounting conservatism.1 powerful empirical evidence that accounting conser-
vatism benefits both lenders and borrowers in the debt
Debtholders’ demand for conservatism contracting process. For lenders, accounting conser-
vatism is more likely to trigger covenant violations and
The separation of ownership and management leads to help lenders reduce downside risk by allowing them to
information asymmetry between debtholders and man- take protective action in a timely manner. For borrowers,
agers (shareholders), which engenders an agency prob- more conservative reporting can reduce lenders’ down-
lem of debt. Managers have incentives to take actions side risk, and lenders would then require lower interest
that potentially transfer wealth from debtholders to rates, which benefits borrowers ex ante. Schipper (2005)
shareholders because of their information advantages. In hypothesises that lenders may ‘protect themselves by
addition, penalising managers ex post is usually too diffi- writing conservative contracts, without requiring biased
cult or costly given their limited liability and tenure. Debt reporting’. Beatty et al. (2008) respond that although
contracting has evolved to alleviate the agency problem debt contracting includes conservative modifications,
of debt (Jensen and Meckling 1976). Accounting num- lenders also demand conservative reporting to mitigate
bers are always used in debt contracting for several rea- agency costs. Both conservative financial reporting

and conservative debt contracting are demanded by metry between shareholders and managers. Managers
debtholders to reduce agency costs. Using a sample have incentives to inflate earnings and withhold bad
consisting of 78,949 firms from 22 countries, Ball et al. news as their compensation is linked to firm perfor-
(2008) argue that debt markets, instead of equity mar- mance. In addition, managers’ tenure and liability are
kets, shape accounting conservatism and highlight the limited. Once a manager leaves a firm, recouping the
importance of the contracting role of financial reporting. losses may be costly or even impossible for sharehold-
Covenants govern the transfer of decision making and ers. Even with a court judgement, shareholders may not
control rights from shareholders to debtholders when a be able to receive sufficient restitution owing to man-
company suffers from financial distress and thereby limit agers’ limited legal liability (Watts 2003a; Kothari et al.
managers’ ability to expropriate debtholders’ wealth. The 2009). Accounting conservatism facilitates shareholders’
use of covenants depends on accounting conservatism. ability to mitigate this agency problem through timely
Covenants are useful only when firms recognise losses loss recognition. Ball and Shivakumar (2005) argue that
in a timely manner. Therefore, when firms’ public debt through timely loss recognition, conservatism reduces
contracting includes more extensive use of covenants, managers’ incentives to invest or continue projects with
their reporting is more conservative (Nikolaev 2010). negative NPV.
However, the institutional environment affects this rela- Another possible explanation is that shareholders
tion. According to Chen et al. (2010), the conservatism want to minimise taxation. Shareholders own residual
of state-owned enterprises is low owing to lenders’ weak claims, indicating that after shareholders pay interest to
demand, and conservatism is negatively related to the debtholders, compensation to managers, taxation to the
fraction of total loans from state-owned banks. Firms government and so on, the residual wealth belongs to
in India report more conservatively in response to the shareholders. Therefore, shareholders have incentives to
entry of foreign banks (Gormley et al. 2012). Investigat- minimise taxation. Conservatism recognises losses in a
ing changes in conservatism after covenant violations, more timely manner than gains, which can defer tax
Tan (2013) finds that financial reporting conservatism payments and improve firm value.
increases immediately after covenant violations. This LaFond and Watts (2008) show that outside equity
phenomenon is more significant when creditors pos- investors demand accounting conservatism owing to in-
sess greater bargaining power, when firms’ operations formation asymmetry between them and firm insiders.
are more volatile and when creditors put Chief Restruc- Conservative information can mitigate this information
turing Officers in place. Using a natural experiment, asymmetry problem because it is reliable and verifi-
Aier et al. (2014) establish a clear causal link between able. Lafond and Roychowdhury (2008) find that firms
debtholders’ demand and accounting conservatism, in- with more severe agency problems between sharehold-
dicating that debtholders’ demand for conservatism ers and managers demand more conservative accounting
leads to accounting conservatism, not vice versa. More- than other firms. In addition, Ramalingegowda and Yu
over, firm creditors with more conservative accounting (2012) find that not all shareholders care about account-
before default have significantly higher recovery rates ing conservatism: only institutional investors who are
than other creditors (Donovan et al. 2015). Empirical likely and able to monitor managers demand more con-
evidence also indicates that debtholders’ demand for servatism. Their empirical evidence shows that higher
conservatism is the most important source of accounting ownership by institutions is associated with more conser-
conservatism. vative financial reporting. Hedge funds are another type
However, Gigler et al. (2009) derive a mathemati- of shareholder that demands accounting conservatism
cal model and show that accounting conservatism de- (Cheng et al. 2015). After hedge fund intervention, firms’
creases the efficiency of debt contracting, which is financial reporting becomes more conservative.
contrary to empirical evidence. Their diverging results
may be explained by their use of a different notion
of efficiency. Investigating the relationship between ac- Auditors’ demand for conservatism
counting conservatism and debt contracting conditional
on liquidation and covenant renegotiation, Li (2013) Auditors demand conservatism because they suffer from
finds that demand for accounting conservatism varies information asymmetry and asymmetric payoffs. Audi-
with the degree of renegotiation costs and liquidation tors are responsible for the reliability and verifiability of
value. financial reporting. When firms are subject to financial
fraud, auditors face reputational harm and risk of prose-
cution by shareholders. Palmrose and Scholz (2000) find
Shareholders’ demand for conservatism that auditors are more likely to be sued when finan-
cial statements are restated. For example, Arthur An-
Executive compensation contracts are used to mitigate dersen faced litigation and became bankrupt after the
the agency problem stemming from information asym- Enron scandal. Time-series variations in conservatism

are consistent with this explanation. Basu (1997) docu- is useful and indispensable for financial reporting (Ball
ments that conservatism increases during periods of in- et al. 2000a; Watts 2003a; Zhang 2008; Ahmed and Du-
creased litigation against auditors, and Holthausen and ellman 2011; Kabir and Laswad 2014); moreover, em-
Watts (2001) confirm this finding. Big Eight auditors pirical evidence indicates that accounting conservatism
report more conservative financial reporting than non- has been increasing up to at least 2002 (Ball and Shiv-
Big Eight auditors, especially when they are exposed to akumar 2006), which is inconsistent with regulators’
greater legal liability (Basu et al. 2001). advocacy for the elimination of conservatism.
Regulators’ demand for conservatism

Managers’ role in accounting conservatism
When firms face bankruptcy, regulators tend to be con-
demned by the media and investors and to even become Managers play a special role in implementing conser-
involved in legal proceedings. Because of the Enron scan- vatism. Accounting principles usually afford managers
dal, the SEC was criticised harshly by investors, and it discretion. For example, accounting principles provide
had to issue the Sarbanes-Oxley Act to protect investors’ managers with several depreciation methods. On one
wealth. One outcome of the 2007–2008 securities mar- hand, managers can use their discretion to produce
ket meltdown is severe criticism of generally accepted high-quality accounting statements that accurately re-
accounting principles (hereafter GAAP), which are pro- flect firms’ economic circumstances. On the other hand,
duced by the FASB (Pinnuck 2012). managers can exploit their discretion to report account-
Researchers (2005; Lambert 2010) and standard set- ing numbers that are in line with their own incentives and
ters argue that conservatism leads to biased financial reduce firms’ value, which is referred to as managerial
reporting and that it is unsuitable for equity valuation. opportunism. For conservative accounting principles,
Consequently, they argue that conservatism should be managers also possess considerable discretion. Managers
eliminated from GAAP. According to the IASB (2008), determine the amount of future cash flows produced by
conservatism will ‘lead to a bias in the reported finan- projects and the time to record assets. Ahmed and Du-
cial position and financial performance’, and ‘this frame ellman (2013) find that over-confident managers tend
work does not include prudence or conservatism as de- to incorporate losses in a less timely manner and tend
sirable qualities of financial reporting information’. to show less conservatism than other managers. These
FASB (2010) in its conceptual framework statement of findings suggest that managerial characteristics affect the
financial accounting concepts No. 8 offers the following level of conservatism in financial reporting.
statement: Certain mechanisms require managers to implement
conservatism even though conservatism is likely to re-
Financial reports represent economic phenomena in duce their compensation and restrict their opportunistic
words and numbers. To be useful, financial informa- behaviour. Specifically, firms’ governance mechanisms
tion not only must represent relevant phenomena, but limit managers’ discretionary behaviours. Debtholders
it also must faithfully represent the phenomena that it and shareholders can foresee managers’ opportunistic
purports to represent. To be a perfectly faithful represen- behaviour and demand conservatism to protect their
tation, a depiction would have three characteristics. It
wealth. If firms do not implement conservatism, both
would be complete, neutral, and free from error . . . Sec-
ond, substance over form, prudence (conservatism), and
debtholders and shareholders will require high capital
verifiability, which were aspects of reliability in Concepts costs (Zhang 2008; Kim et al. 2013). Auditors also have
Statement 2 or the Framework (1989), are not consid- incentives to require managers to implement conser-
ered aspects of faithful representation. Substance over vatism, namely, litigation and reputation risk. In ad-
form and prudence were removed . . . Chapter 3 does dition, litigation against managers forces managers to
not include prudence or conservatism as an aspect of implement conservatism. Chung and Wynn (2008) find
faithful representation because including either would that managerial legal liability coverage, which reduces
be inconsistent with neutrality. Some respondents to the managers’ litigation risk, is negatively related to account-
Discussion Paper and Exposure Draft disagreed with ing conservatism.
that view. They said that the framework should include The outside managerial labour market helps to mon-
conservatism, prudence, or both. They said that bias
itor managers’ opportunistic behaviour. Fama (1980)
should not always be assumed to be undesirable, espe-
cially in circumstances when bias, in their view, produces
argues that analogous to the case of a capital market,
information that is more relevant to some users. the managerial labour market is an efficient market that
properly values managers’ reputation and performance.
Kothari et al. (2010) argue that ‘verifiability and con- If a manager does a good job and creates wealth for
servatism are critical features of a GAAP shaped by mar- shareholders, the managerial labour market will reward
ket forces’. Empirical evidence shows that conservatism the manager through a high valuation. However, if a

manager shirks, pursues his or her own interests and for recognising good news compared with bad news in
runs a company poorly, the manager’s value will decline. financial statements. According to this definition, losses
In addition, lower-level managers will detect and report are recognised in a timely manner with low verification
managerial shirking, which is called ‘internal monitor- and are more likely to be recognised in the current pe-
ing’. All these mechanisms can limit managers’ oppor- riod. Therefore, firms with negative earnings or negative
tunistic behaviour and require managers to implement earnings changes in the current period are more likely to
and maintain conservatism. reverse their situation and have positive earnings in the
next period. Gains require high verification to be recog-
How is Accounting Conservatism nised, and they are more likely to persist to the future un-
Measured? til the related cash flows have been realised. Thus, firms
with positive earnings or positive earnings changes in
Various methods are used to measure accounting con- the current period are more likely to have positive earn-
servatism. Frequently used measures in the literature ings or positive earnings changes in the next period. The
include market-to-book (hereafter MTB), earnings, ac- prediction is tested cross-sectionally by using equation
cruals, accruals/cash flows relation and earnings/stock (1) (the variable definitions are presented in Table 1).
return relation measures. We refer to the MTB measure β1 will be significantly negative if financial reporting is
as a balance sheet measure because it results from the conservative (hereafter the Persistence measure).
understatement of net assets. The earnings, accruals and
accruals/cash flows relation measures are called income X it /P it−1 = α0 + α1 D + β0 X it−1 /P it−2
statement measures. We review the earnings/stock return
+ β1 D × X it−1 /P it−2 + ξ (1)
relation measure in detail because of its wide use.
Because losses are recognised in a timely manner while

Balance sheet measures2 gains are recognised gradually in the future, the distri-
bution of earnings is more likely to be negatively skewed.
The MTB measure is developed by Feltham and Ohlson Consequently, the difference between earnings skewness
(1995, 1996; Givoly and Hayn 2000). Feltham and and cash flows skewness can be used to measure con-
Ohlson (1996) define conservatism as an asymptotic dif- servatism (Givoly and Hayn 2000; Beatty et al. 2008;
ference between the book value and the market value. Ahmed and Duellman 2013; Aier et al. 2014). Further,
The market value includes expected value and growth the skewness of earnings (cash flow) is calculated by
opportunity, while the book value tends to recognise E[(x − μ)3 /σ3 ] , where μ and σ are the mean and stan-
news with prudence. When the book value is actually dard deviation of earnings (cash flow) over the last five
lower than the market value, namely, when the MTB ra- years or five-year window centred on the observation
tio is greater than 1, financial reporting is regarded as year3 (hereafter the Skewness measure). The variables
conservative. A greater MTB ratio indicates more con- are all deflated by total assets. The skewness of earn-
servative financial reporting. However, Beaver and Ryan ings is negative when financial reporting is conservative,
(2000) argue that the MTB ratio is a noisy measure of whereas the skewness of cash flows is not likely to be
conservatism. They split the BTM ratio into two com- negative. Therefore, the difference in skewness between
ponents: a bias component and a lag component. Bias cash flows and earnings is positive. Greater conservatism
means that the book value is persistently higher (lower) results in a larger skewness difference.
than the market value, such that the BTM ratio is persis-
tently above (below) one. Lag indicates that unexpected
Accruals measures
economic gains (losses) are recognised in the book value
over time rather than immediately, such that the BTM
Without conservatism, accruals will disappear and net
ratio is temporarily lower (higher) than its mean (one
income will converge to cash flows over time. Earnings
in the absence of bias) but tends to approach its mean
are usually recognised on an accrual basis, so the asym-
over time. Beaver and Ryan (2000) state that it is only
metric timeliness of earnings affects accruals. Timely
bias components that can be used to measure accounting
loss recognition and gradual gains recognition result in
conservatism.
negative net accruals. Therefore, the level and rate of cu-
mulative negative net accruals can be used to estimate
Income statement measures conservatism (Givoly and Hayn 2000; Beatty et al. 2008;
Ahmed and Duellman 2013; Aier et al. 2014). Accru-
Earnings measures als measures at the firm-year level are equal to averaged
(non-operating) accruals scaled by total assets over the
Basu (1997) defines conservatism to capture accoun- previous five years (with a minimum of two years) or
tants’ tendency to require a higher degree of verification centred on the observation year, multiplied by negative

Table 1 Variable definitions

Variable Definition
X earnings before extraordinary items
P the market value of equity at the beginning of the fiscal year
ACC equals Inventory + Accounts receivable + Other current assets − Accounts payable − Other current liabilities −
Depreciation
CFO cash flows from operations, which equals X less ACC
R (adjust) returns from 9 months before the fiscal year-end to 3 months after the fiscal year-end or (adjust) annual returns
MTB the market value of equity scaled by the book value of equity
Size logarithmic transformation of total assets
Lev defined as long-term and short-term debt deflated by the market value of equity (Khan and Watts 2009)†
D takes the value of 1 when X is negative and 0 otherwise
DCFO equals 1 if CFO is negative and 0 otherwise
DR equals 1 if R is negative and 0 otherwise
x portion of the total unexpected return R that invariably is contemporaneously captured in accounting income X
y portion of the total unexpected return R that is not contemporaneously captured in X unless required by accounting
conservatism
g portion of the total unexpected return R that never is contemporaneously captured in X, but always is incorporated with a
lag
w an indicator variable that takes the value of 1 when conservative accounting rules and practice lead to the recognition of y
in the current period
ξ ‘noise’ in accounting earnings that reverses in the next period
† Some scholars, such as Ahmed and Duellman (2013) and Erkens et al. (2014), use different definitions of Lev.
one (hereafter the NONACC measure). More conserva- β3 is significantly positive (hereafter the ACC/CFO re-
tive earnings results in higher NONACC. lation measure). The variable definitions are given in
Table 1.
The accruals/cash flows relation measure ACC t = β0 + β1 DCFO t + β2 CFOt

+β3 DCFOt × CFOt + νt (2)
Ball and Shivakumar (2005) propose the accruals/cash
flows relation measure to estimate accounting conser-
vatism in private firms in the United Kingdom (UK)
because of their lack of returns. Accruals improve the Earnings/stock return relation measures
ability of earnings to measure firm performance by mit-
igating the negative serial correlation of cash flows; thus, Original model
accruals are contemporaneously negatively related to op-
erating cash flows (Dechow 1994; Dechow et al. 1998). Basu (1997) proposes the earnings/stock return relation
Therefore, β2 in equation (2) will be negative. Ball and measure, which is by far the most widely used measure
Shivakumar (2005) account for conservatism and mod- in conservatism research.4 He interprets conservatism
ify the model to measure accounting conservatism. They to capture accountants’ tendency to require a higher de-
argue that accruals are positively but asymmetrically re- gree of verification for recognising good news compared
lated to current operating cash flows because of conser- with bad news in financial statements, implying that
vatism. For example, revisions to current cash flows for earnings incorporate ‘bad news’ in a more timely man-
a project or asset are positively related to revisions to ner than ‘good news’. To estimate conservatism, he refers
future expected cash flows. Revisions to future expected to negative and positive stock returns as ‘bad news’ and
cash flows are recognised in current earnings by accruals, ‘good news’, respectively. Since Ball and Brown (1968)
implying that accruals are positively related to revisions identified the phenomenon of ‘price leading earnings’,
to expected cash flows. Hence, accruals are positively re- numerous papers have attempted to explain it (Beaver
lated to current cash flows. Economic losses are more et al. 1980; Kothari and Sloan 1992). The phenomenon
likely to be recognised in the current period, while eco- suggests that stock returns reflect not only earnings in-
nomic gains tend to be recognised gradually. The posi- formation but also information from other sources. The
tive relation between accruals and current cash flows will efficient market hypothesis implies that stock returns
thus be greater in the case of losses. To measure conser- can fully reflect different kinds of information over time.
vatism, Ball and Shivakumar (2005) include a dummy Therefore, it is reasonable to use negative stock returns
that equals one if CF O t is negative and zero otherwise in and positive returns to proxy for ‘bad news’ and ‘good
equation (2). Financial reporting is conservatism when news’, respectively. Dietrich et al. (2007) provide an

Figure 2 Path analysis of the determinants of stock price changes (based on Dietrich et al. 2007)
instructive path analysis of the determinants of stock financial reporting. Financial reporting is also conserva-
price changes (see Figure 2). tive when (β0 + β1 )/β0 is significantly greater than one
Link A (B) indicates that prior-period value changes and when the R-square of the negative sample is higher
(current value changes) are recognised in current earn- than that of the positive sample.
ings. Link C (D) indicates that prior-period value
changes (current value changes) are recognised in cur- A firm-year measure based on Basu’s model
rent non-earnings information. Link E (F) indicates that
current earnings information (non-earnings informa- The DT measure is not a firm-year measure of conser-
tion) is reflected in current stock price changes. Note vatism when it is estimated by using a (pooled) cross-
that current stock price changes reflect current informa- sectional sample. For each firm, the current level of con-
tion only. Dietrich et al. (2007) argue that ‘[a] conser- servatism can be estimated by using time-series data
vative accounting practice would record only a portion for previous years. However, the premise is that the
of a positive current-period value change (value changes level of conservatism is stationary, which seems unre-
in current period> 0) in current earnings via link B. alistic because of firms’ changing characteristics and cir-
The remaining portion of positive value change in this cumstances. Researchers demand a firm-year measure of
period would be recorded in earnings in subsequent pe- conservatism because it is more convenient for them to
riods through link A. Conversely, a negative current- detect time-series and cross-sectional variation in con-
period value change (value changes in current period< servatism. Khan and Watts (2009) design a firm-year
0) would be fully recorded in current earnings through measure of conservatism based on the DT measure.
link B’. Watts (2003a) provides four explanations of con-
Basu (1997) uses a piecewise-linear regression of ac- servatism: a contracting explanation, a litigation
counting income on stock returns to measure con- explanation, a taxation explanation and a regulation
servatism. The variable definitions are presented in explanation. Khan and Watts (2009) argue that these
Table 1. four explanations are closely related to firms’ investment
opportunity set (hereafter IOS) (Smith and Watts 1992).
X it /P it = α0 + α1 DRit They use MTB ratio, size and leverage to estimate con-
+ β0 R it + β1 R it × DRit + τit (3) servatism because these three firm-level characteristics
are related to conservatism through IOS (see Figure 3).
Based on their theoretical analysis and Basu’s (1997)
where the subscripts t and i refer to time periods and
model, Khan and Watts (2009) argue that both β0 and
firm, respectively.
β1 in equation (3) are linear functions of the MTB ratio,
β0 measures the sensitivity of earnings to positive
size and leverage.
returns, and (β0 + β1 ) measures the sensitivity of earn-
ings to negative returns. Under conservatism, earnings
incorporate ‘bad news’ in a more timely manner than G Score = β0 = μ1 + μ2 Size i + μ3 MTB i + μ4 L ev i (4)
‘good news’, so β1 should be significantly positive. β1 is
used to measure conditional conservatism and is usually
called the differential timeliness measure (hereafter the
C Score = β1 = λ1 + λ2 Size i + λ3 MTB i + λ4 L ev i (5)
DT measure). Greater β1 indicates more conservative

ings and returns partially cancel each other out. Givoly

et al. (2007) argue that ‘the use of aggregated data to es-
timate the DT measure reduces the power of the tests to
detect conservatism when it is present’. The second char-
acteristic is ‘the nature of economic events’. Some events
do not affect earnings because they are not reportable
in the period when they occur. Roychowdhury and
Watts (2007) refer to these events as ‘rents’ that do not
represent the value of separable net assets.6 However,
these events actually affect returns in the current period.
Therefore, these events will affect the validity of the DT
measure. The last characteristic is ‘disclosure policy’. Al-
Figure 3 The theoretical analysis of C score (based on Khan though the market efficiency is of the semistrong form,
and Watts 2009) managers could affect the release time of information.
Because returns are influenced by the availability of in-
By substituting equation (4) and (5) into equation (3) formation to the public, the DT measure is biased owing
and adding control variables, we get: to the release time of information.
Dietrich et al. (2007) investigate the econometric
properties of the DT measure and show that the DT
X i /P i = α0 + α1 DRi + (μ1 + μ2 Size i
measure is subject to sample-variance-ratio (SVR) bias
+ μ3 MTBi + μ4 Lev i )R i (λ1 + λ2 Size i and sample truncation (ST) bias.7 Their empirical evi-
dence indicates that even using reporting earnings with-
+ λ3 MTBi + λ4 Lev i )R i × DRi + (δ1 Size i
out asymmetric timeliness the DT measure and R-square
+ δ2 MTBi + δ3 Lev i + δ4 DR i Size i will show conservatism. They conclude that both the DT
measure and R-square are biased measures of conser-
+ δ5 DRi MTBi + δ6 itDR i Lev i ) + τi (6)
vatism. Patatoukas and Thomas (2011) provide more
evidence of bias in the DT measure. They claim that
the measure is affected by two empirical regularities: (1)
In equation (6), μit and λit are estimated by using the scale is positively related to deflated mean earnings
annual cross-sectional regressions (Brown et al. 2011). (called a loss effect); and (2) the scale is negatively related
Together with Size it , MTB it and Lev it , we can obtain the to variance in stock returns (called a return variance ef-
C Score and G Score for each firm and year. G Score and fect). Even in the data without conservatism, whether
C Score vary across firms through cross-sectional varia- simulated or actual, the DT measure will show con-
tion in the firm-year characteristics (Size, MTB and Lev) servatism because of these two empirical regularities.
and vary across time through intertemporal variation in Therefore, researchers should be cautious when using
μt and λt (Brown et al. 2011). Therefore, C_Score is a the DT measure or even avoid using the method to mea-
firm-year measure of conservatism. sure conservatism.
Criticisms of the DT measure

Responses to criticisms
Since Basu (1997) proposed the DT measure, it has been
widely used by numerous researchers.5 However, some Roychowdhury and Watts (2007) propose a framework
researchers argue that the DT measure suffers from con- to explain why the MTB ratio is negatively related to
siderable biases and recommend avoiding it (Dietrich the DT measure. They decompose equity value into four
et al. 2007; Givoly et al. 2007; Patatoukas and Thomas components: rents, unverifiable increases in the value
2011). of separable net assets (UNA), verifiable and recognised
Givoly et al. (2007) point out three characteristics of increases in the value of separable net assets (RNA) and
the information environment that affect the validity of net assets at historical cost (NAH). Rents do not repre-
the DT measure. The first characteristic is an ‘aggrega- sent the value of separable net assets, and they are not
tion’ effect. The DT measure is estimated by aggregate reportable in current earnings even though they actually
earnings and aggregate returns because it is difficult or affect current returns. In this framework, the asymmet-
impossible for us to detect individual economic events ric timeliness of earnings principle is applied only to net
and their impact on earnings and returns. However, the assets values (NAV), and it produces a reported book
aggregate indicator will lose some information (Fan and value of the book value of net assets (NAB). Thus, the
Zhang 2012). For example, both good news and bad news difference between NAB and NAV reflects the cumula-
arise in the estimate period, and their effects on earn- tive effect of past asymmetric timeliness and constitutes

the measure of conservatism in the framework (NAC).

However, the DT measure uses stock returns, which
include rents, to measure ‘news’. Hence, the DT mea-
sure is biased in short periods. Roychowdhury and
Watts (2007) show that the DT measure is more ef-
ficient when the measure is estimated over multiple
periods. Figure 4 How accounting income incorporates economic
In an earnings restatement setting, Ettredge et al. income (based on Ball et al. 2013a)
(2012) confirm that both the DT measure and C Score
can effectively measure conservatism. When firms over- (2000a) assume that X t can be expressed as a linear
state their earnings, the level of conservatism in firms’ function of R t .
financial reporting declines. According to the empiri-
cal evidence, firms that overstate their earnings show X it = β0 + β1 R it + τit (9)
less conservatism than control firms during the earnings
overstatement period; moreover, such firms show less where the subscript i refers to firms.
conservatism during the earnings overstatement period The R-square of regression equation (9) represents
than during the post-overstatement period. Therefore, the timeliness of accounting income. When conditional
the DT measure and C Score are valid measures of con- conservatism is taken into consideration and scaled X t
servatism. with P t , equation (9) is modified as equation (3). Using
By deriving a model characterising the means by which the model, Ball et al. (2013a) explain the SVR and ST
accounting income incorporates different types of infor- biases (Dietrich et al. 2007) and conclude that the DT
mation with different lags and noise, Ball et al. (2013a) measure is a valid measure of conditional conservatism.
provide a rigorous analysis from a model incorporating Patatoukas and Thomas (2011) provide more em-
economic income into accounting income to understand pirical evidence indicating that the DT measure is bi-
conditional conservatism. Conservatism, the accounting ased owing to a ‘loss effect’ and a ‘return variance
recognition principle and accounting accruals are fun- effect’. Ball et al. (2013b) provide another interpretation
damental features of the temporal process of incorpo- of this anomalous evidence. Their analysis shows that
rating economic income into accounting income (see the anomalous evidence arises from the lack of controls
Figure 4). Ball et al. (2000a, 2013a) use stock returns for the cross-sectional variation in the expected com-
to represent economic income and employ earnings ponents of returns and earnings, which is usually called
to represent accounting income. The model classifies a correlated omitted variables problem (Greene 2003).
information into four parts. The first kind of informa- To isolate expected returns or earnings in the estima-
tion is incorporated contemporaneously, the second type tion of the DT coefficient, they suggest controlling for
is incorporated contemporaneously or with a lag de- firm characteristics, using an expectation model to con-
pending on its type and magnitude, the third type is trol for expected earnings and expected returns or using
always incorporated with a lag, and the last part repre- unexpected returns and firm fixed effects to control for
sents ‘noise’ that is not reflected in economic income. expected earnings. After these problems are controlled
The model is illustrated below: for, the DT measure remains valid. Therefore, they rec-
ommend that scholars use the DT measure with caution
R t = x t + y t + gt (7) instead of avoiding the measure.
Studies on Accounting Conservatism

X t = x t + w t y t + (1 − w t−1 )y t−1 + g t−1
Since the work of Basu (1997), empirical research of
+ξt − ξt−1 (8) accounting conservatism has flourished over the last
two decades.8 Researchers have attempted to answer the
where the subscript t and t–1 refers to time periods. The following questions. What is accounting conservatism?
variables are defined in Table 1. Who demands conservatism? How do we measure con-
From equations (7) and (8), we can conjecture that servatism? What are the cross-sectional and time-series
these three features affect the way that accounting in- variations in accounting conservatism? In this section,
come incorporates economic income by influencing the we mainly review the cross-sectional and time-series
portion of x t , y t and g t . Conservatism implies that variations in conservatism. The cross-sectional varia-
when y t is related to good news, w t is more likely to tions include the relation between conservatism and
be zero. Otherwise, w t is more likely to be one. Ball et al. corporate governance, the effect of conservatism on

investment efficiency, the information role of conser- ducing managers’ incentives to invest in negative NPV
vatism and the international variation in conservatism. projects ex ante because their compensation is related
to earnings. On the other hand, if a project performs
Accounting conservatism and corporate governance poorly, accounting conservatism requires the losses to
be recognised and the bad news to be transmitted to
outsider investors in a timely manner. Debt contracting
Corporate governance is necessary owing to the conflict-
covenants are then more likely to be violated, and outside
ing interests between the parties to a firm (Jensen and
investors would require managers to stop the poorly per-
Meckling 1976). Shleifer and Vishny (1997) define cor-
forming projects through corporate governance mech-
porate governance as a set of mechanisms to ensure assets
anisms. Therefore, accounting conservatism improves
of firms are used efficiently, guaranteeing the supplier
investment efficiency.9
of finance a return on investment and thus preventing
Firms recognising losses in a timely manner are more
the inappropriate distribution of these assets to man-
likely to have profitable acquisitions and less likely to
agers or other parties at the expense of the rest of the
experience post-acquisition divestitures. More conser-
stakeholders. Corporate governance mechanisms are
vative firms also act more quickly when divestitures oc-
used to align conflicting interests by efficiently mon-
cur (Francis and Martin 2010). This evidence suggests
itoring managers and contracts (Garcı́a Lara et al.
that conservatism encourages managers to make bet-
2009). Conservatism restricts managers’ opportunistic
ter acquisition decisions. Ahmed and Duellman (2011)
behaviour and improves contract efficiency by more
find that more conservative firms have not only higher
rapidly triggering covenant violations based on account-
gross margins and cash flows but also a lower magni-
ing numbers. For example, timely loss recognition not
tude of special item charges after acquisitions. Bushman
only reduces managers’ incentives to overstate earnings
et al. (2011) argue that the sensitivity of investment to
but also forces managers to give up negative NPV projects
declining opportunities increases when firms recognise
and cease poorly performing projects. Therefore, better
losses in a timelier manner. Firms with more conser-
corporate governance encourages managers to imple-
vatism before the 2007–2008 global financial crisis expe-
ment conservatism.
rienced less of a decline in investment activity following
Conservatism is used to mitigate the conflicts over
the onset of the crisis than less conservative firms, and
dividend policy between debtholders and shareholders
the relation between conservatism and investment ac-
(Ahmed et al. 2002). Beekes et al. (2004) document that
tivity is even more profound in firms with financing
firms with a higher proportion of outside board mem-
constraints, greater external financial needs and higher
bers incorporate bad news, but not good news, into earn-
information asymmetry (Balakrishnan et al. 2014). Con-
ings in a timelier manner than other firms. In addition,
servatism improves investment efficiency by mitigating
conservatism is negatively related to the percentage of
managers’ incentives to invest in risky projects; thus,
inside directors but positively related to the percentage
more conservative financial reporting leads to less risky
of outside directors’ shareholding (Ahmed and Duell-
acquisitions when the firms have accounting-based debt
man 2007). Garcı́a Lara et al. (2009) provide further
covenants (Kravet 2014).
empirical evidence on the relationship between con-
servatism and corporate governance. Their results in-
dicate that corporate governance leads to conservatism,
The information role of conservatism
not vice versa. Goh and Li (2011) compare the level of
conservatism between firms with and without material
Information asymmetry between equity investors and
weaknesses in their internal control quality and find that
firm insiders reduces firm value by increasing the cap-
firms with lower internal control quality exhibit lower
ital cost of equity and agency costs (LaFond and Watts
conservatism.
2008). Easley and O’hara (2004) provide theoretical and
empirical evidence that information asymmetry is posi-
Accounting conservatism and investment efficiency tively related to equilibrium required returns. Managers
tend to maximise their own interests by manipulating
Managers tend to overinvest in negative NPV projects financial reporting and investing in inefficient projects,
and continue poorly performing projects particularly which result in deadweight losses (Jensen and Meckling
when firms have abundant cash flows (Francis and Mar- 1976). Conservatism mitigates information asymmetry
tin 2010). Accounting conservatism improves invest- via at least the following two potential mechanisms.
ment efficiency by affecting managers’ incentive to in- On one hand, conservatism requires high verifiability
vest ex ante and by forcing them to abandon poorly for recognising gains, which curbs managers’ incentives
performing projects ex post (Watts 2003a; Ball and Shiv- to overstate earnings, while requiring faster recognition
akumar 2005). On one hand, accounting conservatism for losses, which reduces managers’ incentives to with-
forces managers to employ timely loss recognition, re- hold bad news. Conservatism thus forces managers to

Figure 5 How institutional structure shapes properties of accounting earnings via parties
of firm (based on Bushman and Piotroski 2006)
provide hard (i.e., more reliable) information to out- Ball et al. (2000a) argue that financial reporting qual-
siders. On the other hand, the hard financial reporting ity is more likely to be shaped by customs and insti-
information induces investors to use soft information tutions than by the adopted accounting standards. They
from other resources to evaluate firms’ current and fu- study how the extent of political influence on accounting
ture performance. Therefore, conservatism plays an in- affects the timeliness and conservatism of earnings. The
formation role in the capital market and improves firms’ extent of political influence, at both the country and
stock prices (LaFond and Watts 2008). firm level, is high in code law countries (e.g., France,
Wittenberg-Moerman (2008) observe that accounting Germany, Japan) in which information asymmetry is re-
conservatism, especially timely loss recognition, mit- duced by private communication since stakeholders are
igates information asymmetry in the secondary loan usually present within the corporate governance struc-
market in the United States (US), not only in the equity ture. By contrast, information asymmetry in common
market. By mitigating information asymmetry, account- law countries (e.g., Australia, Canada, UK, US) is mit-
ing conservatism reduces the frequency and timeliness igated by public disclosure, and the extent of political
of management forecasts (Hui et al. 2009). Kim et al. influence is lower than that in code law countries. Consis-
(2013) investigate the impact of conservatism on the eq- tent with their predictions, the empirical evidence shows
uity market and find that more conservative firms suffer that the timeliness of earnings and conservatism in code
from fewer negative returns during announcements of law countries is lower than that in common law coun-
seasoned equity offerings by mitigating the negative im- tries. Ball et al. (2000a) also investigate variations in con-
pact of information asymmetry. Erkens et al. (2014) ar- servatism among common law countries and find that
gue that lender monitoring mitigates information asym- such variations can be explained by differences in liti-
metry and reduces the demand for conservatism. They gation, regulation and taxation. Although the account-
identify an ‘affiliated banker on board’ (hereafter AFB) as ing standards of Hong Kong, Malaysia, Singapore and
a valid monitor mechanism and find that the presence of Thailand are mainly influenced by common law, the level
an AFB reduces the demand for conservatism. Although of conservatism in these countries is not higher than that
lender monitoring can reduce information asymmetry, in code law countries because the preparers’ incentives
conservatism is nevertheless a cost-effective choice. are similar to those in code law countries (Ball et al.
2003). Ball et al. (2003) argue that financial reporting
quality is influenced more by preparers’ incentives than
International variation in conservatism
by accounting standards per se.
Bushman and Piotroski (2006) provide further quan-
An increasing number of articles study the change in
titative evidence about the influence of the institutional
earnings quality stemming from the switch from na-
structure on the properties of accounting earnings (see
tional accounting standards to either international fi-
Figure 5). They argue that the institutional structure,
nancial reporting standards or US GAAP (Bartov et al.
which includes the legal/judicial regime, securities laws,
2005; Barth et al. 2008; Kabir et al. 2010). Research on
the political economy and tax regimes, affects the incen-
international conservatism has flourished since the work
tives of corporate executives, investors, regulators and
of Basu (1997) (Pope and Walker 1999; Ball et al. 2000a,
other participants. They further argue that these incen-
2000b, 2008, 2003; Ball and Shivakumar 2005; Bush-
tives shape the properties of financial reporting through
man and Piotroski 2006; Guay 2006). Because most po-
a complex interplay of accounting standards, legal sys-
litical and economic influences on financial reporting
tem, market, regulation, political pressure and reporting
practices remain local and different, conservatism varies
discretion exercised by managers. Using a large sam-
across countries adopting the same accounting standards
ple of 70 countries, Kanagaretnam et al. (2014) argue
with conservative provisions (Ball 2006; Burgstahler et al.
that national culture affects accounting conservatism
2006; Soderstrom and Sun 2007; Morais and Curto
in the banking industry. They use individualism and
2009).

Figure 6 The time-series variation of auditor litigation risk (based on Kothari et al. 1988)
Table 2 Summary for the articles of the reviewa

Categoryb Articles Main findings
Corporate governance Ahmed and Duellman (2007); Garcı́a Lara et al. Corporate governance improves
(2009); Goh and Li (2011) accounting conservatism.
Earnings management Chen et al. (2007); Jackson and Liu (2010); Accounting conservatism and earnings
Gao (2013) management are neither entirely mutually
exclusive nor entirely one phenomenon.
However, scholars draw inconsistent
conclusions regarding the effect of
conservatism on earning management.
Investment efficiency Francis and Martin (2010); Ahmed and Duellman Accounting conservatism improves
(2011); Bushman et al. (2011); Kravet (2014) investment efficiency and reduces
investment risk.
Information role LaFond and Watts (2008); Wittenberg-Moerman Accounting conservatism mitigates
(2008); Hui et al. (2009); Kim et al. (2013); Erkens information asymmetry.
et al. (2014)
International variation Pope and Walker (1999); Ball et al. (2000a); Ball Accounting conservatism varies with the
et al. (2003); Bushman and Piotroski (2006); institutional environment and national
Kanagaretnam et al. (2014) culture.
Measures Givoly and Hayn (2000); Ball and Shivakumar (2005); Scholars propose several approaches to
Ball and Shivakumar (2006); Givoly et al. (2007); measure accounting conservatism, and
Roychowdhury and Watts (2007); Dietrich et al. one article recommends using multiple
(2007); Khan and Watts (2009); Patatoukas and measures.
Thomas (2011); Ettredge et al. (2012); Ball et al.
(2013a); Ball et al. (2013b); Lawrence et al. (2013);
Collins et al. (2014)
Explanation Ahmed et al. (2002); Beaver and Ryan (2005); Qiang Empirical evidence shows that
(2007); Beatty et al. (2008); Ball et al. (2008); Chung debtholders, shareholders, auditors,
and Wynn (2008); Zhang (2008); LaFond and suppliers and customers demand
Roychowdhury (2008); Suijs (2008); Gigler et al. accounting conservatism. Debtholders
(2009); Chen et al. (2010); Nikolaev (2010); Brown compose the main party demanding
et al. (2011); Hui et al. (2012); Ramalingegowda and conservatism.
Yu (2012); Gormley et al.(2012); Tan (2013); Ahmed
and Duellman (2013); Aier et al. (2014); Kabir and
Laswad (2014); Cheng et al. (2015); Donovan et al.
(2015)
a This table includes articles from Accounting Review, Journal of Accounting Research, Journal of Accounting and Economics, Contemporary
Accounting Research, Review of Accounting Studies, Australian Accounting Review and Journal of Business Finance & Accounting.
b All articles are classified into seven categories. Articles on corporate governance discuss the relation between accounting conservatism and
corporate governance. Articles on earnings management discuss the relation between accounting conservatism and earning management.
Articles on explanations of conservatism answer what leads to accounting conservatism. Articles on the informational role of conservatism
describe the role of accounting conservatism in mitigating information asymmetry. Articles on international variation in conservatism
show the variation in conservatism between different markets. Articles on investment efficiency investigate the impact of accounting
conservatism on investment efficiency. Articles on measures examine how accounting conservatism is measured.
uncertainty avoidance to measure national culture and is generally consistent with the regulation and litiga-
find that individualism (uncertainty avoidance) is nega- tion explanations (Basu 1997; Givoly and Hayn 2000;
tively (positively) related to conservatism. Holthausen and Watts 2001; Ball and Shivakumar 2006).
Basu (1997) investigates the relation between auditor
litigation risk and conservatism from 1963 to 1990 by
Time-series variation in conservatism using the DT measure. He divides the period into four
stages based on the auditor’s litigation risk (see Figure 6)
Scholars have investigated the country-level time-series (Kothari et al. 1988). The empirical evidence shows that
variation in conservatism, and the empirical evidence the DT measure for the periods 1963–1966, 1966–1975,

Table 3 Accounting conservatism articles decomposed by scholars except the VM measure. In all, 22 papers in-
measure and categorya vestigate why accounting conservatism has become one
Number of Number of of the most important properties of financial reporting,
Measuresb articles Categoryc articles while 13 papers examine how accounting conservatism
VM 1 Corporate governance 3
is measured. Relatively fewer papers investigate other
Persistence 5 Earnings management 3 issues.
C score 9 Investment efficiency 4
MTB 9 Information role 5
Skewness 9 International variation 7 Conclusion
ACC/CFO relation 10 Measures 13
NONACC 13 Explanation 22
DT measure 38 Based on this review, at both the firm and country
level, conservatism has long had a significant influence
a This table includes articles from Accounting Review, Journal of on financial reporting and accounting practice. Ster-
Accounting Research, Journal of Accounting and Economics, Con-
temporary Accounting Research, Review of Accounting Studies,
ling (1970) even refers to conservatism as the most in-
Australian Accounting Review and Journal of Business Finance & fluential principle of valuation in accounting. Studies
Accounting. have shown that shareholders, auditors, regulators, sup-
b There are eight measures of accounting conservatism:
pliers, customers and debtholders demand accounting
VM, C_score, MTB, Persistence, Skewness, ACC/CFO relation, conservatism to reduce agency costs by mitigating in-
NONACC, DT. For details, please see ‘How is accounting conser-
vatism measured?’. Note that one article uses at least one mea-
formation asymmetry and facilitating corporate gover-
sure. nance. Together with a firm’s corporate governance, con-
c All the articles are classified into seven categories. Articles on cor- servatism improves investment efficiency and reduces
porate governance discuss the relation between accounting con- investment risk. Data on time-series variation show that
servatism and corporate governance. Articles on earnings man- financial reporting is more conservative until 2002 –
agement discuss the relation between accounting conservatism
and earnings management. Articles on explanations of conser-
despite criticism regarding conservatism. Overall, based
vatism answer what leads to accounting conservatism. Articles on on prior research on conservatism, the conclusion can
the informational role of conservatism describe the role of ac- be drawn that conservatism remains one of the most im-
counting conservatism in mitigating information asymmetry. Arti- portant properties of financial reporting, that it deserves
cles on international variation in conservatism show the variation further research attention and that it cannot be excluded
in conservatism between different markets. Articles on investment
efficiency investigate the impact of accounting conservatism on in-
from accounting standards.
vestment efficiency. Articles on measures examine how accounting Many research opportunities remain for researchers.
conservatism is measured. As far as we know, no paper has carefully investigated
regulators’ demands for conservatism. Recently, the
FASB and IASB excluded conservatism from the joint
1975–1982 and 1983–1990 is 0.009, 0.178, 0.002 and conceptual framework for financial reporting. Research
0.214 respectively, which is consistent with the variation must therefore investigate the regulators’ demand for
in the auditor’s litigation risk (Basu 1997). Holthausen conservatism and emphasise the important role of
and Watts (2001) extend Basu’s (1997) study and investi- conservatism in financial reporting.
gate the variation in conservatism since 1927. They find Ahmed and Duellman (2013) find that managerial
similar results, also consistent with the regulation and overconfidence affects the level of conservatism in
litigation explanations. In addition, they provide weak financial reporting. What about other managerial char-
evidence for the contracting explanation as financial re- acteristics? For instance, their knowledge of accounting
porting exhibits conservatism prior to the increase in and tenure may affect accounting conservatism. Further-
litigation in the late 1960s. more, because auditors must audit financial statements
and issue an opinion before their release, auditor
Summary of empirical research on conservatism characteristics may also affect accounting conservatism.
Conservatism does not come without a cost. It affects
In this section, we summarise the articles on accounting the neutrality of financial reporting, and it is unsuitable
conservatism falling within the scope of the review pub- for equity valuation. An optimal level of conservatism
lished between 1 January 1998 and 31 March 2015 (see exists. Roychowdhury and Martin (2013) argue that
Table 2).10 These articles offer primary insights on ac- non-discretionary conservatism – or, more accurately,
counting conservatism. The articles on accounting con- ‘normal conservatism’ – is the level of conservatism that
servatism can be decomposed by measure and category is determined by the firm’s economic circumstances.
(see Table 3). The DT measure is used by 38 articles and We should develop a more powerful model to estimate
is the most widely used measure among all the measures. ‘normal conservatism’ in order to avoid additional costs
The other measures are used more or less frequently by resulting from excessive conservatism.

Recently, an increasing number of papers has shown Ahmed, A.S. and Duellman, S. 2011, ‘Evidence on the Role
that conservatism affects acquisition efficiency and risk. of Accounting Conservatism in Monitoring Managers’ In-
No papers have investigated the impact of conservatism vestment Decisions’, Accounting & Finance, 51 (3): 609–
on other types of investment. As conservatism consti- 33.
tutes an important property of financial reporting and Ahmed, A.S. and Duellman, S. 2013, ‘Managerial Overcon-
potentially affects other investment efficiency, the con- fidence and Accounting Conservatism’, Journal of Accounting
clusions drawn regarding acquisitions should be extend Research, 51 (1): 1–30.
to other types of investment. Ahmed, A.S., Morton, R.M. and Schaefer, T.F. 2000, ‘Account-
ing Conservatism and the Valuation of Accounting Numbers:
Acknowledgements Evidence on the Feltham-Ohlson (1996) Model’, Journal of Ac-
counting, Auditing & Finance, 15 (3): 271–92.
The authors are grateful for comments on an earlier draft Aier, J.K., Chen, L. and Pevzner, M. 2014, ‘Debtholders’ De-
of the paper from Weiwei Gao (Xi’an Jiaotong Univer- mand for Conservatism: Evidence from Changes in Directors’
sity) and Huaili Lv (Shanghai University). Fiduciary Duties’, Journal of Accounting Research, 52 (5): 993–
1027.
Balakrishnan, K., Watts, R.L. and Zuo, L. 2014, ‘The Effect
Notes
of Accounting Conservatism on Corporate Investment During
the Global Financial Crisis’, MIT Sloan Research Paper No.
1 In addition to these main parties, Hui et al. (2012) argue that
4941-11.
firms’ suppliers and customers also demand conservatism.
2 Balance sheet measures include the MTB measure and a Valuation Ball, R. 2006, ‘International Financial Reporting Standards
Model measure (hereafter VM measure). We do not review the (IFRS): Pros and Cons for Investors’, Accounting & Business
VM measure because of its rare use. For details, please see Ahmed Research, 36: 5–27.
et al. (2000).
3 Some researchers (Beatty et al. 2008; Aier et al. 2014) also calculate Ball, R. and Brown, P. 1968, ‘An Empirical Evaluation of Ac-
Skewness over a 20-quarter window (with a minimum of five counting Income Numbers’, Journal of Accounting Research, 6
quarters). (2): 159–78.
4 This paper has been cited in 2863 published or working papers
as of 1 April 2015, according to Google Scholar. Ball, R., Kothari, S. and Nikolaev, V.V. 2013a, ‘Econometrics of
5 Ryan (2006) provides a good review of papers that use the DT the Basu Asymmetric Timeliness Coefficient and Accounting
measure to identify conditional conservatism. He indicates the Conservatism’, Journal of Accounting Research, 51 (5): 1071–97.
limitations of these papers and provides suggestions to improve
the model. Ball, R., Kothari, S.P. and Nikolaev, V.V. 2013b, ‘On Estimat-
6 A detailed description of rents is provided in Roychowdhury and ing Conditional Conservatism’, The Accounting Review, 88 (3):
Watts (2007). Under this framework, the DT measure is a biased 755–87.
measure of conservatism.
7 For a detailed discussion of SVR and ST bias, see Beaver et al. Ball, R., Kothari, S.P. and Robin, A. 2000a, ‘The Effect of
(1987) and Maddala and Lahiri (1992). International Institutional Factors on Properties of Account-
8 Watts (2003a; 2003b) and Ryan (2006) provide good reviews of ing Earnings’, Journal of Accounting and Economics, 29 (1): 1–
accounting conservatism. 51.
9 Ball and Shivakumar (2005) point out that only conditional
conservatism can improve investment efficiency. In addition, Ball, R., Robin, A. and Sadka, G. 2008, ‘Is Financial Report-
Francis and Martin (2010) argue that conservatism is effec- ing Shaped by Equity Markets or by Debt Markets? An In-
tive only in combination with strong governance and a control ternational Study of Timeliness and Conservatism’, Review of
system. Accounting Studies, 13 (2–3): 168–205.
10 We chose the time point of 1 January 1998, because Basu pub-
lished ‘The conservatism principle and the asymmetric timeliness Ball, R., Robin, A. and Wu, J.S. 2000b, ‘Accounting Standards,
of earnings’ in 1997. the Institutional Environment and Issuer Incentives: Effect on
Timely Loss Recognition in China’, Asia-Pacific Journal of Ac-
counting & Economics, 7 (2): 71–96.
References Ball, R., Robin, A. and Wu, J.S. 2003, ‘Incentives versus Stan-
dards: Properties of Accounting Income in Four East Asian
Ahmed, A.S., Billings, B.K., Morton, R.M. and Stanford-Harris, Countries’, Journal of Accounting and Economics, 36 (1–3): 235–
M. 2002, ‘The Role of Accounting Conservatism in Mitigating 70.
Bondholder-Shareholder Conflicts over Dividend Policy and
in Reducing Debt Costs’, The Accounting Review, 77 (4): 867– Ball, R. and Shivakumar, L. 2005, ‘Earnings Quality in UK Pri-
90. vate Firms: Comparative Loss Recognition Timeliness’, Journal
of Accounting and Economics, 39 (1): 83–128.
Ahmed, A.S. and Duellman, S. 2007, ‘Accounting Conser-
vatism and Board of Director Characteristics: An Empirical Ball, R. and Shivakumar, L. 2006, ‘The Role of Accruals in
Analysis’, Journal of Accounting and Economics, 43 (2–3): 411– Asymmetrically Timely Gain and Loss Recognition’, Journal of
37. Accounting Research, 44 (2): 207–42.

Barth, M.E., Landsman, W.R. and Lang, M.H. 2008, ‘Interna- Chandra, U. 2011, ‘Income Conservatism in the US Technology
tional Accounting Standards and Accounting Quality’, Journal Sector’, Accounting Horizons, 25 (2): 285–314.
of Accounting Research, 46 (3): 467–98.
Chen, H., Chen, J.Z., Lobo, G.J. and Wang, Y. 2010, ‘Asso-
Bartov, E., Goldberg, S.R. and Kim, M. 2005, ‘Compar- ciation between Borrower and Lender State Ownership and
ative Value Relevance among German, U.S., and Interna- Accounting Conservatism’, Journal of Accounting Research, 48
tional Accounting Standards: A German Stock Market Per- (5): 973–1014.
spective’, Journal of Accounting, Auditing & Finance, 20 (2): Chen, Q.I., Hemmer, T. and Zhang, Y.U. N. 2007, ‘On the
95–119. Relation between Conservatism in Accounting Standards and
Basu, S. 1995, Conservatism and the Asymmetric Timeliness of Incentives for Earnings Management’, Journal of Accounting
Earnings, PhD, Universtiy of Rochester. Research, 45 (3): 541–65.
Basu, S. 1997, ‘The Conservatism Principle and the Asym- Cheng, C.S.A., Huang, H.H. and Li, Y. 2015, ‘Hedge Fund In-
metric Timeliness of Earnings’, Journal of Accounting and Eco- tervention and Accounting Conservatism’, Contemporary Ac-
nomics, 24 (1): 3–37. counting Research, 32 (1): 392–421.
Basu, S., Hwang, L.-S. and Jan, C.-L. 2001, ‘Differences in Chung, H.H. and Wynn, J.P. 2008, ‘Managerial Legal Liability
Conservatism between Big Eight and Non-Big Eight Auditors’, Coverage and Earnings Conservatism’, Journal of Accounting
Available at: SSRN: http://ssrn.com/abstract=2428836. and Economics, 46 (1): 135–53.
Beatty, A., Weber, J. and Yu, J.J. 2008, ‘Conservatism and Debt’, Collins, D. W., Hribar, P. and Tian, X. 2014, ‘Cash Flow Asym-
Journal of Accounting and Economics, 45 (2–3): 154–74. metry: Causes and Implications for Conditional Conservatism
Research’, Journal of Accounting and Economics, 58 (2–3): 173–
Beaver, W.H., Lambert, R. and Morse, D. 1980, ‘The Infor- 200.
mation Content of Security Prices’, Journal of Accounting and
Economics, 2: 3–28. Dechow, P.M. 1994, ‘Accounting Earnings and Cash Flows
as Measures of Firm Performance: The Role of Accounting
Beaver, W.H., Lambert, R.A. and Ryan, S.G. 1987, ‘The Infor- Accruals’, Journal of Accounting and Economics, 18 (1): 3–
mation Content of Security Prices: A Second Look’, Journal of 42.
Accounting and Economics, 9 (2): 139–57.
Dechow, P.M., Kothari, S.P. and Watts, R. 1998, ‘The Relation
Beaver, W.H. and Ryan, S. 2000, ‘Biases and Lags in Book Value between Earnings and Cash Flows’, Journal of Accounting and
and their Effects on the Ability of the Book-to-Market Ratio to Economics, 25 (2): 133–68.
Predict Book Return on Equity’, Journal of Accounting Research,
38 (1): 127–48. Devine, C.T. 1963, ‘The Rule of Conservatism Reexamined’,
Journal of Accounting Research, 1: 127–38.
Beaver, W.H. and Ryan, S.G. 2005, ‘Conditional and Uncon-
ditional Conservatism: Concepts and Modeling’, Review of Ac- Dietrich, J.R., Muller, K., III and Riedl, E. 2007, ‘Asymmetric
counting Studies, 10 (2–3): 269–309. Timeliness Tests of Accounting Conservatism’, Review of Ac-
counting Studies, 12 (1): 95–124.
Beekes, W., Pope, P. and Young, S. 2004, ‘The Link between
Earnings Timeliness, Earnings Conservatism and Board Com- Donovan, J., Frankel, R. and Martin, X. 2015, ‘Accounting Con-
position: Evidence from the UK’, Corporate Governance: An servatism and Creditor Recovery Rate’, The Accounting Review,
International Review, 12 (1): 47–59. 90 (6): 2267–303.
Bliss, J.H. 1924, Management Through Accounts, The Ronald Easley, D. and O’Hara, M. 2004, ‘Information and the Cost of
Press Co., New York. Capital’, The Journal of Finance, 59 (4): 1553–83.
Brown, P., Dobbie, G.W. and Jackson, A.B. 2011, ‘Measures of Erkens, D.H., Subramanyam, K.R. and Zhang, J. 2014, ‘Af-
the Timeliness of Earnings’, Australian Accounting Review, 21 filiated Banker on Board and Conservative Accounting’, The
(3): 222–34. Accounting Review, 89 (5): 1703–28.
Burgstahler, D.C., Hail, L. and Leuz, C. 2006, ‘The Importance Ettredge, M., Huang, Y. and Zhang, W. 2012, ‘Earnings Re-
of Reporting Incentives: Earnings Management in European statements and Differential Timeliness of Accounting Con-
Private and Public Firms’, The Accounting Review, 81 (5): 983– servatism’, Journal of Accounting and Economics, 53 (3): 489–
1016. 503.
Bushman, R.M. and Piotroski, J.D. 2006, ‘Financial Reporting Fama, E.F. 1980, ‘Agency Problems and the Theory of the Firm’,
Incentives for Conservative Accounting: The Influence of Legal Journal of Political Economy, 88 (2): 288–307.
and Political Institutions’, Journal of Accounting and Economics,
42 (1–2): 107–48. Fan, Q. and Zhang, X.-J. 2012, ‘Accounting Conservatism, Ag-
gregation, and Information Quality’, Contemporary Accounting
Bushman, R.M., Piotroski, J.D. and Smith, A.J. 2011, ‘Capital Research, 29 (1): 38–56.
Allocation and Timely Accounting Recognition of Economic
Losses’, Journal of Business Finance & Accounting, 38 (1–2): Feltham, G.A. and Ohlson, J.A. 1995, ‘Valuation and
1–33. Clean Surplus Accounting for Operating and Financial

Activities’, Contemporary Accounting Research, 11 (2): 689– Holthausen, R.W. and Watts, R.L. 2001, ‘The relevance of the
731. Value-relevance Literature for Financial Accounting Standard
Setting’, Journal of Accounting and Economics, 31 (1–3): 3–
Feltham, G.A. and Ohlson, J.A. 1996, ‘Uncertainty Resolution
75.
and the Theory of Depreciation Measurement’, Journal of Ac-
counting Research, 34 (2): 209–34. Hui, K.W., Klasa, S. and Yeung, P.E. 2012, ‘Corporate Suppli-
ers and Customers and Accounting Conservatism’, Journal of
Financial Accounting Standards Board (FASB) 1975. ‘Account-
Accounting and Economics, 53 (1–2): 115–35.
ing for Contingencies’, Statement of Financial Accounting Stan-
dards No. 5, FASB, Norwalk, CT. Hui, K.W., Matsunaga, S. and Morse, D. 2009, ‘The Impact of
Conservatism on Management Earnings Forecasts’, Journal of
Financial Accounting Standards Board (FASB) 1989, ‘Qualita-
Accounting and Economics, 47 (3): 192–207.
tive Characteristics of Accounting Information’, Statement of
Financial Accounting Concepts No. 2, FASB, Norwalk, CT. International Accounting Standards Board (IASB) 2008, ‘An
Financial Accounting Standards Board (FASB) 2009. ‘Intangi- Improved Conceptual Framework for Financial Reporting:
bles – Goodwill and Other’, Accounting Standards Codification Chapter 1, The Objective of Financial Reporting, and Chap-
350, FASB, Norwalk, CT. ter 3, Qualitative Characteristics and Constraints of Decision-
Useful Financial Reporting Information’, Exposure Draft May
Financial Accounting Standards Board (FASB) 2010, ‘Concep- 2008, IASB, London.
tual Framework for Financial Reporting: Chapter 1, The Ob-
jective of General Purpose Financial Reporting, and Chapter Jackson, S.B. and Liu, X. 2010, ‘The Allowance for Uncollectible
3, Qualitative Characteristics of Useful Financial Reporting In- Accounts, Conservatism, and Earnings Management’, Journal
formation’, Statement of Financial Accounting Concepts No. 8, of Accounting Research, 48 (3): 565–601.
FASB, Norwalk, CT. Jensen, M.C. and Meckling, W.H. 1976, ‘Theory of the Firm:
Francis, J.R. and Martin, X. 2010, ‘Acquisition Profitability and Managerial Behavior, Agency Costs and Ownership Structure’,
Timely Loss Recognition’, Journal of Accounting and Economics, Journal of Financial Economics, 3 (4): 305–60.
49 (1): 161–78. Ji, K. 2013, ‘Better Late Than Never, the Timing of Goodwill
Gao, P. 2013, ‘A Measurement Approach to Conservatism and Impairment Testing in Australia’, Australian Accounting Review,
Earnings Management’, Journal of Accounting and Economics, 23 (4): 369–79.
55 (2–3): 251–68. Kabir, M.H. and Laswad, F. 2014, ‘The Behaviour of Earnings,
Garcı́a Lara, J., Garcı́a Osma, B. and Penalva, F. 2009, ‘Ac- Accruals and Impairment Losses of Failed New Zealand Fi-
counting Conservatism and Corporate Governance’, Review of nance Companies’, Australian Accounting Review, 24 (3): 262–
Accounting Studies, 14 (1): 161–201. 75.
Gigler, F., Kanodia, C., Sapra, H. and Venugopalan, R. 2009, Kabir, M.H., Laswad, F. and Islam, M.A. 2010, ‘Impact of IFRS
‘Accounting Conservatism and the Efficiency of Debt Con- in New Zealand on Accounts and Earnings Quality’, Australian
tracts’, Journal of Accounting Research, 47 (3): 767–97. Accounting Review, 20 (4): 343–57.
Givoly, D. and Hayn, C. 2000, ‘The Changing Time-series Prop- Kanagaretnam, K., Chee Yeow, L. and Lobo, G.J. 2014, ‘In-
erties of Earnings, Cash Flows and Accruals: Has Financial fluence of National Culture on Accounting Conservatism and
Reporting Become More Conservative?’, Journal of Accounting Risk-taking in the Banking Industry’, The Accounting Review,
and Economics, 29 (3): 287–320. 89 (3): 1115–49.
Givoly, D., Hayn, C.K. and Natarajan, A. 2007, ‘Measuring Khan, M. and Watts, R.L. 2009, ‘Estimation and Empirical
Reporting Conservatism’, The Accounting Review, 82 (1): 65– Properties of a Firm-year Measure of Accounting Conser-
106. vatism’, Journal of Accounting and Economics, 48 (2–3): 132–
50.
Goh, B.W. and Li, D. 2011, ‘Internal Controls and Conditional
Conservatism’, The Accounting Review, 86 (3): 975–1005. Kim, Y., Li, S., Pan, C. and Zuo, L. 2013, ‘The Role of Accounting
Conservatism in the Equity Market: Evidence from Seasoned
Gormley, T.A., Kim, B.H. and Martin, X. 2012, ‘Do Firms Equity Offerings’, The Accounting Review, 88 (4): 1327–56.
Adjust Their Timely Loss Recognition in Response to Changes
in the Banking Industry?’, Journal of Accounting Research, 50 Kothari, S.P., Lys, T.Z., Watts, R.L. and Smith, C.W. 1988, ‘Audi-
(1): 159–96. tor Liability and Information Disclosure’, Journal of Accounting,
Auditing & Finance, 3: 307–39.
Greene, W. 2003, Economitric Analysis, Pearson Education, Up-
per Saddle River. Kothari, S.P., Ramanna, K. and Skinner, D.J. 2010, ‘Implica-
tions for GAAP from an Analysis of Positive Research in Ac-
Guay, W. 2006, ‘Discussion of the Role of Accruals in Asymmet- counting’, Journal of Accounting and Economics, 50 (2–3): 246–
rically Timely Gain and Loss Recognition’, Journal of Accounting 86.
Research, 44 (2): 243–55.
Kothari, S.P., Shu, S. and Wysocki, P.D. 2009, ‘Do Managers
Hanna, J.D. 2002, ‘Conservatism and Accounting Measure- Withhold Bad News?’, Journal of Accounting Research, 47 (1):
ment’, Working Paper, University of Chicago. 241–76.

Kothari, S.P. and Sloan, R. 1992, ‘Information in Prices About Roychowdhury, S. and Martin, X. 2013, ‘Understanding Dis-
Future Earnings: Implications for Earnings Response Coeffi- cretion in Conservatism: An Alternative Viewpoint’, Journal of
cients’, Journal of Accounting and Economics, 15 (2–3): 143–71. Accounting and Economics, 56 (2–3, Supplement 1): 134–46.
Kravet, T.D. 2014, ‘Accounting Conservatism and Managerial Roychowdhury, S. and Watts, R.L. 2007, ‘Asymmetric Timeli-
Risk-taking: Corporate Acquisitions’, Journal of Accounting and ness of Earnings, Market-to-book and Conservatism in Finan-
Economics, 57 (2–3): 218–40. cial Reporting’, Journal of Accounting and Economics, 44 (1):
2–31.
Lafond, R. and Roychowdhury, S. 2008, ‘Managerial Own-
ership and Accounting Conservatism’, Journal of Accounting Ryan, S.G. 2006, ‘Identifying Conditional Conservatism’, Eu-
Research, 46 (1): 101–35. ropean Accounting Review, 15 (4): 511–25.
LaFond, R. and Watts, R.L. 2008, ‘The Information Role of Schipper, K. 2005, ‘Fair Values in Financial Reporting’. Avail-
Conservatism’, The Accounting Review, 83 (2): 447–78. able at: http//fars.org/2005AAAFairValueKSchipper.pdf, ac-
cessed 10 April 2013.
Lambert, R. 2010, ‘Discussion of “Implications for GAAP from
an Analysis of Positive Research in Accounting”’, Journal of Scott, W.R. 2012, Financial Accounting Theory, Prentice-Hall,
Accounting and Economics, 50 (2–3): 287–95. Toronto.
Lawrence, A., Sloan, R. and Sun, Y. 2013, ‘Non-discretionary Shleifer, A. and Vishny, R.W. 1997, ‘A Survey of Corporate
Conservatism: Evidence and Implications’, Journal of Account- Governance’, Journal of Finance, 52 (2): 737–83.
ing and Economics, 56 (2–3, Supplement 1): 112–33. Smith, C.J. and Watts, R.L. 1992, ‘The Investment Opportunity
Set and Corporate Financing, Dividend, and Compensation
Li, J. 2013, ‘Accounting Conservatism and Debt Contracts: Effi-
Policies’, Journal of Financial Economics, 32 (3): 263–92.
cient Liquidation and Covenant Renegotiation’, Contemporary
Accounting Research, 30 (3): 1082–98. Soderstrom, N.S. and Sun, K.J. 2007, ‘IFRS Adoption and Ac-
counting Quality: A Review’, European Accounting Review, 16
Maddala, G.S. and Lahiri, K. 1992, Introduction to Economet-
(4): 675–702.
rics, Macmillan, New York.
Sterling, R.R. 1970, The Theory of the Measurement of Enterprise
Mather, P. 1999, ‘Financial Covenants in Australian Bank-loan
Income, University of Kansas Press, Lawrence, KS.
Contracts: Incidence, Measurement Rules and Monitoring’,
Australian Accounting Review, 9 (17): 63–72. Suijs, J. 2008, ‘On the Value Relevance of Asymmetric Finan-
cial Reporting Policies’, Journal of Accounting Research, 46 (5):
Morais, A.I. and Curto, J.D. 2009, ‘Mandatory Adoption of 1297–321.
IASB Standards: Value Relevance and Country-specific Fac-
tors’, Australian Accounting Review, 19 (2): 128–43. Tan, L. 2013, ‘Creditor Control Rights, State of Nature Ver-
ification, and Financial Reporting Conservatism’, Journal of
Nikolaev, V.V. 2010, ‘Debt Covenants and Accounting Conser- Accounting and Economics, 55 (1): 1–22.
vatism’, Journal of Accounting Research, 48 (1): 51–89.
Watts, R.L. 2003a, ‘Conservatism in Accounting Part I: Expla-
Palmrose, Z.-V. and Scholz, S. 2000, ‘Restated Financial State- nations and Implications’, Accounting Horizons, 17 (3): 207–
ments and Auditor Litigation’, Working Paper. Available at: 21.
http://dx.doi.org/10.2139/ssrn.248455, accessed 10 April 2013.
Watts, R.L. 2003b, ‘Conservatism in Accounting Part II: Ev-
Patatoukas, P.N. and Thomas, J.K. 2011, ‘More Evidence of Bias idence and Research Opportunities’, Accounting Horizons, 17
in the Differential Timeliness Measure of Conditional Conser- (4): 287–301.
vatism’, Accounting Review, 86 (5): 1765–93.
Watts, R.L. and Zimmerman, J.L. 1986, Positive Accounting
Pinnuck, M. 2012, ‘A Review of the Role of Financial Reporting Theory, Prentice-Hall, Toronto.
in the Global Financial Crisis’, Australian Accounting Review,
22 (1): 1–14. Wittenberg-Moerman, R. 2008, ‘The Role of Information
Asymmetry and Financial Reporting Quality in Debt Trad-
Pope, P.F. and Walker, M. 1999, ‘International Differences in ing: Evidence from the Secondary Loan Market’, Journal of
the Timeliness, Conservatism, and Classification of Earnings’, Accounting and Economics, 46 (2–3): 240–60.
Journal of Accounting Research, 37 (3): 53–87.
Wyatt, A., Matolcsy, Z. and Stokes, D. 2001, ‘Capitalisation
Qiang, X. 2007, ‘The Effects of Contracting, Litigation, Regu-
of Intangibles – a Review of Current Practice and the Reg-
lation, and Tax Costs on Conditional and Unconditional Con-
ulatory Framework’, Australian Accounting Review, 11 (24):
servatism: Cross-sectional Evidence at the Firm Level’, The Ac-
22–38.
counting Review, 82 (3): 759–96.
Ramalingegowda, S. and Yu, Y. 2012, ‘Institutional Ownership Zhang, J. 2008, ‘The Contracting Benefits of Accounting Con-
and Conservatism’, Journal of Accounting and Economics, 53 servatism to Lenders and Borrowers’, Journal of Accounting and
(1–2): 98–114. Economics, 45 (1): 27–54.


Accounting Conservatism: A Literature Review: Yuxiang Zhong, Wanli Li

Uploaded by

Copyright:

Available Formats

Accounting Conservatism: A Literature Review: Yuxiang Zhong, Wanli Li

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

Accounting Conservatism: A Literature Review: Yuxiang Zhong, Wanli Li

Uploaded by

Copyright:

Available Formats

Accounting Conservatism: A Literature Review

Yuxiang Zhong, School of Management, Xi’an Jiaotong University

A important properties of financial reporting

Australian Accounting Review No. 0 Vol. 0 Issue 0 2016 doi: 10.1111/auar.12107 1

Figure 1 The conceptual framework of the review

2 Australian Accounting Review

4 Australian Accounting Review

Regulators’ demand for conservatism

6 Australian Accounting Review

Because losses are recognised in a timely manner while

Table 1 Variable definitions

The accruals/cash flows relation measure ACC t = β0 + β1 DCFO t + β2 CFOt

8 Australian Accounting Review

ings and returns partially cancel each other out. Givoly

Criticisms of the DT measure

10 Australian Accounting Review

the measure of conservatism in the framework (NAC).

Studies on Accounting Conservatism

12 Australian Accounting Review

Table 2 Summary for the articles of the reviewa

14 Australian Accounting Review

16 Australian Accounting Review

18 Australian Accounting Review

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.