providers require independent audits, the fact that they do implies that capital providers consider that

auditors increase credibility. Audited financial statements are generally accepted as giving credibility to
the annual report in which they are found. Intuitively, qualified audit reports should be more valued by
market participants. However, they do not provide timely signals to capital markets, basically because
the qualification has been expected.33
In contrast, the work of financial intermediaries adds value in the capital market because they use
and interpret accounting data so that share prices reflect the result of their research.34 Analysts’ earnings
forecasts are more accurate than time‐series models of earnings. Their accuracy is affected by innate
ability, company assignments, brokerage affiliation and industry specialisation. Their earnings forecasts
and recommendations affect share prices. However, analysts’ forecasts have been found to be biased in
an overly optimistic direction, especially when their brokerage house has been hired to underwrite par-
ticular security issues.35 Interestingly, research shows that firms with more intangible assets attract more
analysts who expend more effort to follow them.36

Voluntary disclosure theory

According to Core, the voluntary disclosure literature offers the best opportunity for increasing under-
standing of the role of accounting information in firm valuation and corporate finance.37 The most
important questions relate to which firms voluntarily disclose, how stakeholders use this information in
allocating capital to firms and how the various people who produce or use this information verify, regu-
late and interpret it. Although the literature has theorised about voluntary disclosure, empirical evidence
from reputable studies is scant.
Voluntary disclosure theory predicts that shareholders optimise disclosure policy, corporate gover-
nance and management incentives to maximise firm value.38 Increased disclosure lowers information
asymmetry, thus lowering the cost of capital. However, this has to be weighed against the costs associ-
ated with incentives, litigation and proprietorship. Increased disclosure does not mean credible or unbi-
ased disclosure because it is too costly to eliminate all the ways in which managers can cause bias. The
theory predicts that even though disclosure is somewhat biased, on average it will be credible. Although
manipulation of disclosures is possible, corporate governance is designed to constrain managers to
follow the optimal policy. Firms will differ in their disclosures according to their disclosure policy and
the ability of the individual firm’s governance to enforce the policy.
Managers with shares or share options may take actions to manipulate the prices of their share or
option holdings.39 They delay disclosing good news and speed up the release of bad news before their
stock option award periods to maximise their stock‐based compensation. Managers may find it more
profitable to buy shares before their stock options are awarded.40 Management forecasts are associated
with trading by insiders in the firm’s shares, and with share option compensation that is at risk because
of firm performance. Firms with greater information asymmetry use more share and option incentives.
Greater information asymmetry is associated with more voluntary disclosure, because managers do not
want to bear any risk associated with any misvaluation of shares. Firms that warn investors that bad
news is imminent have significantly more negative returns per unit of unexpected earnings than firms
that do not warn, suggesting that firms are penalised for disclosing bad news early.41

8.5 Value relevance

LEARNING OBJECTIVE 8.5 Explain the role of information and information intermediaries in capital
markets.
Recall that association studies regress accounting numbers on capital market data to test for signifi-
cant relationships to assess the value relevance of information. Value relevance research has two major
features.42 First, this research requires in‐depth knowledge of accounting institutions, accounting stan-
dards and reported accounting numbers. Second, timeliness is not an important issue as it is in an event
study. An item of accounting information is considered relevant if it has the ability to make a difference

CHAPTER 8 Capital market research and accounting  245

to the decisions of financial statement users. An accounting number is value relevant if it conveys infor-
mation that results in the modification of investors’ expectations about future payoffs or associated
risk.43 Value relevance research attempts to assess the relevance and reliability of accounting amounts
because accounting’s Conceptual Framework stresses the importance of accounting information being
both relevant and faithfully represented. An accounting amount is faithfully represented if it represents
what it purports to represent. Value relevance research tests relevance and faithful representation jointly
as it is difficult to test these characteristics separately. Figure 8.2 depicts graphically the two concepts.
Relevance relates to two aspects of the underlying economics of the investment: a value construct of
some kind (such as expenditure on acquisition of goodwill), and the process by which value is expected
to be created. Faithful representation refers to two links: the relevance link and a measure that is capable
of reflecting the economic substance of the value construct and process. Faithful representation is
affected by accounting rules (GAAP), economic uncertainty — especially if it causes the link from the
value construct to value creation to be ill‐defined — and management discretion to communicate cred-
ibly with investors.44

FIGURE 8.2 The relationship between relevance and reliability, and value relevance

Relevance Faithful representation Value relevance

Value
Value
creation Measure Value
construct
process

1. Affected by uncertainty about link from value

construct to value creation (e.g. purchased goodwill)
2. Affected by GAAP (e.g. full expensing of research and development)
3. Affected by management discretion
Source: Adapted from A Wyatt.45

Value relevance research does not attempt to assess the usefulness of accounting numbers. Nor can it
be inferred from the statistical test alone that the information of interest causes the level of market value,
changes in share prices, or financial performance. Value relevance tests provide a statistical association
only, an association that is not backed up by theory of the underlying links between accounting, standard
setting and value.46
Studies have focused on accounting information by examining four different associations:
1. earnings and security returns
2. the value relevance of non‐earnings data
3. the value relevance of different accounting practices
4. the value relevance of different GAAP.
Studies examining these associations use three different research methods as follows.
1. Relative association studies compare the association between stock market values or changes in those
values and alternative bottom‐line accounting measures. The accounting number with the greater
R2 is described as being more value relevant. R2, known as the coefficient of determination, is a
measure used in statistical model analysis to assess how accurately a model explains and predicts
future outcomes.
2. Incremental association studies investigate whether the accounting number of interest is helpful in
explaining value or returns over a long period. The accounting number is said to be value relevant if
its estimated regression coefficient is significantly different from zero.

246  Contemporary issues in accounting

3. Marginal information content studies investigate whether a particular accounting number adds to the
information set available to investors. Typically, these are event studies examining whether the release
of an accounting number is associated with value changes. Price changes are considered evidence of
value relevance. These studies represent less than 10% of the value relevance studies.47
Note that these studies explore an association. An association signifies that when two or more sets of
numbers are regressed and an R2 is given, which suggests a relationship between the numbers. However,
the use of the term ‘association’ signifies the lack of theory to explain why the sets of numbers are
related.
Holthausen and Watts found two implicit but different theories in the literature.48 The first assumes that
accounting earnings tend to be highly associated with equity market value changes. In the second theory,
accounting’s role is to provide information on inputs to valuation models that investors use in valuing firms’
equity. Most researchers assume that the main role of financial reporting is to provide measures associated
with value or measures of value, or information relevant for equity valuation, an assumption criticised by
Holthausen and Watts. As noted in previous chapters, accounting information is an important input into
other contexts, not just capital markets, and is used by many non‐investor user groups.
Holthausen and Watts identify three assumptions underlying the value relevance literature.
1. Equity investors are assumed to be the main or dominant users of financial reports. They are assumed
to use those reports mainly for the valuation of equity.
2. Share prices adequately represent investors’ use of information in valuing equity securities.
3. Share‐price‐based tests of relevance and reliability measure relevance and reliability as defined by FASB
statements (reflecting the heavy dominance of American researchers in stock market research).49

8.6 What value relevance studies tell us

LEARNING OBJECTIVE 8.6 Distinguish between behavioural finance findings and mainstream finance
findings.
Studies of the relationship between accounting earnings and share returns have analysed data from most
of the world’s major stock exchanges. Results show that any evident relationship is weak at best.
Reported earnings are not good measures of the value‐relevant events that are built into share prices in
the periods studied. The relatively low association between reported earnings and share prices suggests
that earnings do not capture all the information incorporated into share prices.
One reason for this may be that investors focus on all events that affect future cash flows, while earn-
ings only capture those events that meet the criteria for accounting recognition. Nichols and Wahlen dis-
agree.50 They report that annual earnings changes contain more value relevant information than changes
in cash flows. Other reasons suggested for the low association between reported earnings and share
prices refer to managers’ risk preferences and negative earnings. Risk‐averse managers are likely to
report more con­servative earnings figures than less risk‐averse ones and are likely to report bad news
earlier. Both of these actions have been confirmed. Another explanation may be that negative earnings
are not value relevant because investors do not expect losses to persist, but firms with losses are included
in test samples, lowering the association. Additionally, losses do not provide information about a firm’s
ability to generate future cash flows.

Relevance and faithful representation

Value relevance research examines the association between accounting amounts and equity market
values.51 An accounting amount is considered value relevant if it has an association with share prices
although the literature uses the term equity market values. As mentioned, value relevance research
focuses on the relevance and faithful representation criteria that are used in the Conceptual Framework
to choose among accounting methods. The assumption is that an accounting amount will be value rel-
evant only if the amount reflects information relevant to investors in valuing the firm and if the amount
is measured reliably enough to be reflected in share prices.

CHAPTER 8 Capital market research and accounting  247

 Some studies have tried to decompose earnings and test the association of decomposed items
with share prices. The evidence is conflicting. Some studies show that partitioning earnings into pre‐­
exceptional, exceptional and extraordinary improves the association, while others contradict these find-
ings, especially in relation to extraordinary earnings. Earnings are more correlated with share returns
than cash flows for short periods, but with increased time intervals both tend to have the same level of
correlation with returns. Inclusion of non‐earnings variables increases the correlation between returns
and accounting data.52
Consolidated earnings were found to have an incremental information content beyond that of the
parent company, so that consolidation improves the value relevance of earnings. However, no value rel-
evance was found for the minority interest portion of earnings and equity.
For Finnish investors at least, restating local GAAP earnings to conform to international accounting
standards has helped meet foreign investor needs but appears to have been of limited use for domestic
investors. Asset revaluations under Australian GAAP, on the other hand, are relevant and estimated with
some reliability but are not considered timely.53 The degree of association for value relevance of earn-
ings differs internationally. The degree of association is lower in countries with bank‐oriented rather than
market‐oriented economies, in countries in which private‐sector bodies are not involved in the standard‐
setting process and in continental Europe. Differences, especially between the United Kingdom and the
United States, seem to be sensitive to the earnings measure analysed.54

Measurement perspective research

Fair value accounting is a focus of value relevance research. Much of the US‐based research has
focused on banks because most of their assets and liabilities are financial. That research suggests that
for investment securities, loans and derivatives, fair values are incrementally informative, relative to
their book values, in explaining bank share prices.55 Other research has focused on pension and other
post‐retirement assets and liabilities. These are perceived by investors as assets and liabilities of the
firm, although they are less reliably measured than other recognised assets and liabilities.56 The fair
value of pension assets measures the pension asset implicit in share prices more reliably than the book
value of those pension assets. Other findings are that disaggregated costs are potentially more infor-
mative to investors than aggregated costs. Investors perceive the fair value estimates of debt, equity
securities and bank loans as more relevant than historical cost amounts. As well, investors perceive
fair values of derivatives as reflecting more precision than their notional amounts in relation to the
underlying economic value.
The costs of intangible assets are relevant to investors and are reflected in share prices with some relia-
bility. Investors perceive expenditures on research, development and advertising as capital acquisitions.
As well, capitalised software, patents and goodwill are found to be relevant to investors. In contrast, fair
value estimates of tangible long‐lived assets are not value relevant, although this may be because these
values are not always reliably estimated. Because Australian and British accounting standards permitted
revaluation, some of the research focused on the revaluation of tangible long‐lived assets. Revaluations
were found to have incremental explanatory power relative to earnings and changes in earnings and
markets found revaluation estimates made by external appraisers more informative and reliable than
those made by internal appraisers such as directors. Evidence of whether fair value measurements are
likely to be used by firms facing financial distress is contradictory. Relations between revaluations and
share prices are weaker for firms with high debt‐to‐equity ratios, suggesting that managers of these
firms manipulate the earnings numbers via asset revaluation. However, Danish bank regulators have
used mark‐to‐market accounting for Danish banks for a long time and there is no reliable evidence
that Danish banks manage mark‐to‐market to avoid regulatory constraints.57 Landsman also reports that
US‐based evidence on share options suggests that managers facing incentives to manage earnings are
likely to do so when fair values of share options must be estimated using entity‐supplied estimates for
the options.

248  Contemporary issues in accounting

 Landsman concludes from the evidence relating to fair values and value relevance that fair values are
informative to investors but the level of informativeness is affected by measurement error and the source
of the estimates of fair value (external appraisers or management).58 The evidence also suggests that
investors are provided with information that is somewhat faithfully represented and relevant.

The efficiency of capital markets

Capital markets are assumed to be efficient. The hypothesis suggests that if accounting choices and
changes do not affect cash flows, investors will not alter their assessment of share prices. Therefore,
market efficiency is important to accounting because the assumption means that investors would see
through alternative or opportunistic accounting choices. If markets are not efficient, then discretionary
accruals (or earnings management) can fool investors.
Researchers have had little success in resolving the question of whether markets are efficient in
relation to discretionary or cosmetic accounting choices. Tests for earnings management generate results
with little explanatory power for earnings management.59 Dechow and Skinner argued that it is sufficient
in well‐functioning markets for information to be disclosed, because rational investors will process the
information appropriately.60 Not all empirical evidence is consistent with this opinion.61

Testing whether capital markets are efficient

Market efficiency is important to accounting because this assumption means that investors would see
through alternative or opportunistic accounting choices. If markets are not efficient, then discretionary
accruals can fool investors. Tests of market efficiency in the late 1970s and 1980s began to under-
mine the efficient market hypothesis, although the research methods employed were unable to determine
whether investors could see through cosmetic changes or mandated accounting changes.62
Most short‐window event studies are generally consistent with market efficiency although sometimes
the market does not react quickly to information, so that there is a ‘drift’, which contradicts market
efficiency. Longer‐period tests assume that the market can over- or under- react to new information
because of human judgement and behavioural biases. Recent evidence, such as large abnormal returns
spread over several years after well‐publicised events such as initial public offerings, also contradict
market efficiency. According to Kothari, collectively this research poses a formidable challenge to the
efficient market hypothesis.63
Evidence in studies examining manipulation of discretionary accruals immediately before initial
public offerings and other equity offerings also challenges market efficiency. These studies suggest that
the market fails to recognise the optimistic bias in earnings forecasts, even though owners, managers and
analysts have incentives to issue overly optimistic forecasts.64
Statistical anomalies continue to appear in studies using the capital asset pricing model (CAPM). Investors
were shown not to react ‘logically’ to new information.65 In 1992, Eugene Fama, a key figure in the develop-
ment of the CAPM model, withdrew his support from it. Defenders of the efficient market hypothesis began to
argue that noisy data, measurement errors and selection bias could explain anomalies. Fields, Lys and Vincent
note that the evidence about market efficiency is not conclusive on whether markets are inefficient or not.66
Both before and after the global financial crisis, researchers, finding evidence of market inefficiency, were
beginning to draw on the behavioural literature for support, arguing that investors are not necessarily rational.
Prices in an efficient market theoretically reflect all that there is to know about capital assets. This
ideology endorsed markets as a perfect allocative device. However, as discussed, anomalies (information
that conflicts with the efficient market hypothesis) appeared. The main anomalies are as follows.
•• The small firm effect. CAPM understates cross‐sectional returns of listed firms with low market values
of equity and overstates those of firms with high market values of equity.
•• The neglect effect. Returns of firms not followed by analysts are inferior to those of firms followed by
many analysts.

CHAPTER 8 Capital market research and accounting  249

•• The exchange effect. Investor interest and publicly available information vary according to the market
in which a firm’s shares are traded.
•• The exotic effect. There are end‐of‐month, end‐of‐year, weekend, Yom Kippur and January effects.
For example, many of the abnormal returns for smaller firms occur during the first half of January
and stock returns are predictably negative over weekends. These results cannot be explained by the
efficient market hypothesis and CAPM combination.
Contemporary issue 8.2 discusses the failure of the efficient market hypothesis and other mathemat-
ical models in predicting the global financial crisis.

8.2 CONTEMPORARY ISSUE

Mathematical models
In relation to the global financial crisis, the popular
belief is that conventional economics and finance have
failed because their models failed to predict the global
financial crisis, to prevent the global financial crisis and
even caused the global financial crisis. What seemed
to fail in particular were the efficient market hypothesis
(EMH) and the widespread use of mathematical finan-
cial models.
Did the mathematical models fail? Is mathematics
the curse of economics and finance? Investment banks
and hedge funds hired well‐qualified mathematicians
(‘quants’) to help them understand and model the markets using complex mathematics. Around 2000,
new mathematical models were invented that made it easier to price collateralised debt obligations
(CDOs). One of these models was the Gaussian Copula Function, devised by a quant (David X Li)
working at JPMorgan. The formula allowed determination of the correlation between the default rates of
different securities. As one commentator put it, if this model was correct, it would tell you the likelihood
that related CDOs would explode, as well as the likelihood that a given set of corporations would default
on their bond debt in quick succession. Various commentators have said that the formula will go down
in history as the instrument that brought the world financial system to its knees. Li is unlikely to get a
Nobel Prize as was believed when his formula was adopted.
As profit margins on CDOs narrowed, subprime housing loans and other lesser quality loans were
brought into the CDOs. Then the market started doing things that the model had not expected — a
model that had not been extensively tested by those putting it into use. Events underlying the CDOs
were not independent or random, but complex and difficult to analyse. Li’s formula had oversimplified
things, not recognising that there could be correlations between random events because of factors such
as employer linkages, geographic regions or acts of God. Despite not understanding it, many adopted
the model, assuming it was accurate.
In relation to other models also blamed for the global financial crisis, Paul Krugman, the Nobel Prize
winning economist, said that economists mistook beauty, clad in impressive looking mathematics, for
truth. Warren Buffet warned his shareholders to ‘beware of geeks bearing formulas’. Others have said
that the desire for elegant mathematical models plays down the role of bad behaviours. The quants are
not entirely to blame — some blame must rest with those who bought the instruments that they created.
Source: Based on information from JE King, ‘Economists and the global financial crisis’, Global Change, Peace &
Security; Steve Keen, ‘Was the GFC a mathematical error?’, Business Spectator; Damien Wintour, ‘The equation that
sank Wall Street?’, Necessary and Sufficient.67

QUESTIONS
1. If mathematical models were abandoned, what could replace them?
2. Discuss whether mathematical models are better to be partly right rather than totally wrong.
3. Should decision makers be more mathematically literate so that they understand the limitations of the
models they use or should the models be extensively tested before use?

250  Contemporary issues in accounting

8.7 Behavioural finance
LEARNING OBJECTIVE 8.7 Discuss how behavioural research contributes to an understanding of
decision making.
The global financial crisis of 2008 increased uncertainty over whether capital markets are efficient. As
a consequence, behavioural finance has become a topic of considerable interest. Although the origins of
behavioural finance date to 1951, about the same time as modern finance was being born, interest in it
did not gain momentum until the late 1980s. This renewed interest seems to have been engendered by
two developments:
1. mounting empirical evidence suggesting that existing finance theories appear to be deficient in
fundamental ways — in particular, theories as to why individual investors trade, how they perform,
how they choose their portfolios and why returns vary across shares for reasons other than risk68
2. the development of prospect theory by Kahneman and Tversky. This theory is based on the
simple idea that the pain associated with a given amount of loss (say $100) is greater than the
pleasure derived from an equivalent gain, so that investors attach more importance to avoiding
the loss.69 The underlying assumptions of this alternative model of decision making are more realistic
than those of existing finance theories.70
Contemporary issue 8.3 reviews the anomaly of black swan effects in relation to investor behaviour.

8.3 CONTEMPORARY ISSUE

Black swan events

In 1697 the Dutch discovered black swans in Western
Australian. Until then all swans were thought to be
white. Centuries after the Dutch discovery the term
‘black swan event’ was popularised, indicating an
event thought impossible but does actually happen.
Black swan events are now considered events that
come as a surprise, have a major impact or conse-
quence and are only rationalised with the benefit
of hindsight. Like the discovery of black swans they
are rare, beyond the realm of expectation and there-
fore not predictable or predicted. A finance professor,
writer and former Wall Street trader, Nassim Nicholas
Taleb, popularised the term in relation to the economy and capital markets when he described unex-
pected events that had serious consequences. The events do not have to be negative but they do have
a profound effect both on the economy and the markets they disrupt. Taleb argues that it is important
that people assume that a black swan event is possible.
Usually, black swan events in financial markets are meant to refer to panics characterised by massive
waves of selling. Asset prices formerly thought to be uncorrelated fall en masse and liquidity evap-
orates. As part of the global financial crisis, the financial crash of the United States housing market in
2008 was an example of a black swan event. The global financial crisis began in July 2007 with a credit
crunch and led to a loss of confidence by US investors in the value of subprime housing mortgages,
which in turn caused a liquidity crisis. This resulted in the US Federal Bank injecting a large amount of
capital into financial markets. By September 2008, the crisis had worsened as stock markets around
the globe crashed. The effect of the crash was catastrophic. Taleb used the 2008 financial crisis and the
idea of black swan events to point out if a broken system is allowed to fail, it will strengthen the system
against the catastrophe of future black swan events.
Similar to the 2008 financial crisis, the dot‐com bubble of 2001 is another example of a black swan
event. Since the internet was at its infancy in terms of commercial use, various investment funds were
investing in technology companies with inflated valuations. When these companies failed, the invest-
ment funds were hit hard, with the downside being passed on to their investors. The digital industry was
new so that it was nearly impossible to predict the collapse.

CHAPTER 8 Capital market research and accounting  251

 Black swan events illustrate the limits of theory, the bounds of prediction and the fallibility of assump-
tions. On this basis, the expression is ideally suited to financial markets because they are unpredictable.
Because they are hard to predict they will probably cause a major shift in the perceptions of inves-
tors. With black swan events in the recent past, could others be lurking? Answers are hard to come by.
Neither market analysts nor media outlets are likely to be reliable guides for when and where potential
black swans might emerge again.
Commentators suggest that having strong convictions about shares or the stock market in the long
term is surely more foolish than believing something on a shorter time frame. The longer the time frame,
the more guesswork. They also suggest that thinking in terms of a long‐term promise or making a long‐
term judgement about a share on the basis of the available evidence is irrational.
In the end, not all swans are white. When shares are purchased, the research, assumptions and
expectations behind that decision are already history.
Sources: Based on information from Andrew Osterland, ‘Fight or flight? Threat of black swan events spooks investors’, CNBC;
Marcus Padley, ‘Truth about long‐term investment and black swans’, The Sydney Morning Herald; ‘Black swan’, Investopedia.71

QUESTIONS
1. What is a black swan event?
2. How do black swan events illustrate the limits of theory?
3. Why was the global financial crisis labelled a black swan event?
4. What advice can be drawn from financial market black swan events?

In their work, Kahneman and Tversky integrated psychology and economics, providing the intellec-
tual foundations of behavioural finance. Their focus was decision making under uncertainty, a charac-
teristic of capital markets. They demonstrated that decision making involves the use of heuristics and
systematically departs from the laws of probability. Modern finance involves little or no examination of
individual decision making. Deduction is prominent, so that decision making is a ‘black box’. Because
finance is concerned with prediction rather than description or explanation, finance theorists constructed
abstractions of the decision process.72
Investment decisions are characterised by high exogenous uncertainty because future performance
must be estimated from a set of noisy and vague variables. Investors who make decisions have an
intuitive, less quantitative, emotionally driven perception of risk than that implied by finance models.
Decision makers’ preferences tend to be multifaceted, easily changed and often only formed during the
decision‐making process. They seek satisfactory rather than optimal solutions. The typical investor can
be termed homo heuristics, not homo economics, a completely rational decision maker focused on utility
or wealth maximisation.73

Cornerstones of behavioural finance

An important cornerstone of behavioural finance is cognitive psychology. Because cognitive psychology
is the study of how people perceive, speak, think, remember, or solve problems, it suggests the following.
•• People make systematic errors in the way they think. They use heuristics or rules of thumb to make
decision making easier, which can lead to biases and sub‐optimal investment decisions.
•• People are overconfident about their abilities. Men are more overconfident than women. Entrepreneurs
are especially likely to be overconfident.
•• People put too much weight on recent experience so that they underweigh long‐term averages.
•• Mental accounting separates decisions that should be combined.
•• Framing says that how a concept is presented to people matters. This refers to the old adage about
whether a glass is half full or half empty and how each gives us a different perception about the
quantity in the glass.
•• People avoid realising paper losses but seek to realise paper gains. This behaviour is called the
disposition effect.

252  Contemporary issues in accounting

•• Anchoring says that people tend to rely on a numerical anchor value that is explicitly or implicitly
presented to them and use it as an initial starting point. When things change, people tend to be slow
to pick up on the changes as well as to underreact because of their conservatism. Any evaluation of
returns is distorted by the size of the anchor.
•• Representativeness says that people tend to rely on stereotypes — for example, past performances
are extrapolated without considering the exogenous uncertainty and randomness of financial markets.
Good brand image and high brand awareness result in a lower perception of investment risk.
•• Affect heuristic indicates that emotions affect risk–return perceptions and investment behaviour.
Positive emotional associations result in lower perceived investment risk.
Whether expertise moderates these outcomes is uncertain. Some findings show that investor exper-
tise has no influence on the use of heuristics.74 Other evidence suggests that individual knowledge has a
moderating effect.
These observations contradict the core theories of modern finance, which assume that:
•• investors are perfectly rational (or markets act as if they were)
•• markets are efficient
•• transaction costs are so small that informed traders quickly notice and take advantage of mispricing,
driving prices back to ‘proper’ levels.75
Behavioural finance argues that investors, based on these observations, are not rational, so that there
are observable biases. The list of biases is growing and includes:
•• overconfidence — the tendency of investors to overestimate their skills.
•• endowment effect — the tendency of individuals to insist on a higher price for something they wish to
sell, rather than to buy the same item if they do not already own it.
•• loss aversion — the tendency for people to be risk averse in relation to profit opportunities but to be
willing to gamble to avoid a loss.
•• anchoring — the tendency for people to make decisions based on an initial estimate that is later
adjusted, but not sufficiently adjusted to eliminate the influence of the initial estimate.
•• framing — the tendency to make different choices based on how the decision is framed, especially if
it is framed in terms of a likelihood of a good outcome or the reciprocal bad outcome.
•• hindsight — the tendency to read the present into assessments of the past.
Those who study capital markets are becoming increasingly disillusioned with the assumptions that
underlie the notion that markets are efficient. So what are the implications for accounting of markets not
being efficient and of behavioural finance? Accounting policy makers have relied on market efficiency to
make choices between accounting methods. How will the tenets of behavioural finance affect accounting
policy choice? This is one issue that relies on a ‘wait and see’ answer.

CHAPTER 8 Capital market research and accounting  253