BEHAVIOURAL FINANCE: A BACKROUND BRIEFING

Professor Richard G.P. McMahon,

Head, School of Commerce,

The Flinders University of South Australia,

GPO Box 2100,

Adelaide South Australia 5001.

Telephone: +61 8 82012840

Facsimile: +61 8 82012644

Email: Richard.McMahon@flinders.edu.au

SCHOOL OF COMMERCE RESEARCH PAPER SERIES: 05-9 ISSN: 1441-3906

Acknowledgments:

My thanks are due to Associate Professor Carol Tilt and Mr Matthew Tilling of the School of

Commerce at Flinders University for consistently demonstrating to me the value of

approaching one’s field of scholarship with an open and inquiring mind.

2

BEHAVIOURAL FINANCE: A BACKGROUND BRIEFING

Abstract

The principal purpose of this paper is to provide a background briefing on behavioural

finance (BF) for those unfamiliar with this significant paradigm shift underway in

finance scholarship. The paper synthesises and summarises a rapidly burgeoning

literature on BF which, presently, is not as accessible as that concerned with modern

finance theory. Attention is first given to what is meant by the rationality of financial

agents, and the possibility is introduced that financial agents in an uncertain real world

may be less than strictly rational. Thereafter, the various heuristics and cognitive biases

that may characterise financial decision-making in an uncertain real world are

catalogued and explained. The paper finishes with an assessment of the potential for BF

to transform scholarship in finance.

The economist may attempt to ignore psychology, but it is sheer impossibility for him to

ignore human nature. . . . If the economist borrows his conception of man from the

psychologist, his constructive work may have some chance of remaining purely economic in

character. But if he does not, he will not thereby avoid psychology. Rather, he will force

himself to make his own, and it will be bad psychology.

Clark (1918, p. 4)

Introduction

Modern Finance Theory (MFT), as presented in textbooks and taught in universities around

the world, has evolved over more than 50 years. Seminal works in the earlier stages of this

evolution include Dean (1951) on capital budgeting, Markowitz (1952, 1959) on portfolio

theory, Modigliani and Miller (1958, 1963) on capital structure decisions, Miller and

Modigliani (1961) on dividend policy decisions, Sharpe (1964) and Lintner (1965a, 1965b) on

capital asset pricing, Fama (1970) on capital market efficiency, Black and Scholes (1973) on

option pricing, Jensen and Meckling (1976) on agency theory, Ross (1976) on arbitrage

3

pricing theory, and Leland and Pyle (1977) on signalling theory. MFT is in great part

normative and, primarily for reasons of tractability, much of it rests on strong assumptions

regarding perfect capital markets and the strict rationality of financial agents.

The theoretical advances identified in the previous paragraph, some of which ultimately

resulted in Nobel prizes for their proponents, were followed by a burgeoning of empirical

studies intended to test their validity in a less than perfect world. By the end of the 1970s,

there emerged a growing dissatisfaction with the state of MFT. For example, so inconclusive

had become theoretical and empirical perspectives on such significant elements of MFT as

capital structure decisions and dividend policy decisions that they were styled ‘puzzles’ by

leading finance scholars (Black, 1976; Myers, 1984). Perhaps reflecting a collective

escalation of commitment bias and/or status quo bias, theoreticians nevertheless tended to

dismiss challenges to the explanatory power of their elegant models as mere anomalies that

further research would eventually resolve.

Notwithstanding such defences, a widely-held belief emerged that ‘finance consists of

theories for which there is no evidence and empirical facts for which there is no theory’ (De

Bondt and Thaler, 1995, p. 386). Drawing attention to the essence of the dilemma, De Bondt

and Thaler (1995, p. 387) go on (emphasis added):

. . . the problems with modern finance theory are created by its presumed dual

purpose, characterizing optimal choice and describing actual choice. The validity

of the theory for the first purpose is not in question. However, since it is assumed

that actual people do optimize (or behave as if they did), the theories are also

thought to be good descriptive models. Of course, if people fail to optimize, this is

not the case. The solution is to retain the normative status of optimization (e.g.,

teach students to maximise expected utility and to use Bayes’ rule) but develop

explicitly descriptive models of behavior in markets and organizations. We call

this effort behavioral finance.

4

Thus, Behavioral Finance (BF) emerged in the 1980s amid growing discontent with the then

existing models of MFT, largely stemming from the lack of realism in the assumptions on

human behaviour underpinning these models.

This development did not take place without challenge. For example, in 1986 (p. s466)

Miller wrote:

. . . the rationality-based market equilibrium models in finance . . . are alive and well –

or at least in no worse shape than other comparable models in economics at their level

of aggregation. The framework is not so weighed down with anomalies that a

complete reconstruction (on behavioral/cognitive or other lines) is either needed or

likely to occur in the near future.

Statman (1999, p. 19) expresses the opposing view which did prevail and lead to the creation

of a new paradigm in finance:

I argue, to the contrary, today’s standard finance is so weighted down with anomalies

that reconstructing financial theory along behavioral lines makes much sense.

So far has BF progressed over the last two decades that it now has the customary

accoutrements of an established scholarly field, with the formation of the Institute of

Behavioral Finance in 1998 and the establishment of the Journal of Behavioral Finance in

2000. Detailed reviews of the development and precepts of BF are provided (inter alia) by

Thaler (1992), Thaler (1993), De Bondt and Thaler (1995), Schwartz (1998), Shefrin (1999),

Statman (1999), Schleifer (1999), Hirshleifer (2001), and Barberis and Thaler (2003). The

seminal work in the field is generally recognised to be De Bondt and Thaler (1985).

BF examines financial phenomena through the dual lenses of finance and of cognitive

psychology. However, Statman (1999, p. 19) draws attention to a misconception regarding

BF:

Some people think that behavioral finance introduced psychology into finance, but

psychology was never out of finance. Although models of behavior differ, all

behavior is based on psychology.

5

This, of course, resonates with the quotation from Clark (1918) at the beginning of the paper.

The psychological foundations of BF are provided by the pioneering experimental research on

human decision-making reported by Tversky and Kahneman (1971, 1973, 1974, 1981, 1983,

1986, 1992) and Kahneman and Tversky (1972, 1973, 1979, 1984). The principal thrust of

this and subsequent research is to challenge the strict rationality assumptions of MFT and to

introduce the possibility that, in their decision-making, financial agents employ heuristics and

exhibit certain systematic cognitive biases that together lead to significant departures from the

tenets of MFT.

The principal purpose of this paper is to provide a background briefing on BF for those

unfamiliar with this significant paradigm shift underway in finance scholarship. The paper

synthesises and summarises a rapidly burgeoning literature on BF which, presently, is not as

accessible as that concerned with MFT. Attention is first given to what is meant by the

rationality of financial agents, and the possibility is introduced that financial agents in an

uncertain real world may be less than strictly rational. Thereafter, the various heuristics and

cognitive biases that may characterise financial decision-making in an uncertain real world

are catalogued and explained. The paper finishes with an assessment of the potential for BF to

transform scholarship in finance, especially when it comes to the various puzzles that MFT

has failed to resolve. While a great deal of the literature on BF has been concerned generally

with the functioning of capital markets, and with financial asset pricing in particular, less

attention has been paid to the potential significance of BF for corporate finance. To reveal

some of the possibilities in the latter field, where appropriate the emphasis in this paper is

upon corporate finance applications of BF.

6

Rationality of Financial Agents

Strict Rationality

As indicated in the introduction to this paper, normative MFT rests on strong assumptions

regarding perfect capital markets and the strict rationality of financial agents (a perfect

markets and perfect people perspective). On the rationality of financial agents, De Bondt and

Thaler (1995, p. 389) summarise the position taken as follows:

Although modern finance typically makes predictions about market outcomes and the

behavior of firms, there is an underlying set of assumptions about individual behavior

that are used to derive these predictions. Specifically, people are said to be risk averse

expected utility maximisers and unbiased Bayesian forecasters. In other words, agents

make rational choices based on rational expectations.

The significance of each element of this description is explained below, and then various

defences that have been proposed for this view of financial agents are identified.

So-called cardinal utility theory, as first articulated by Von Neuman and Morgenstern

(1944), begins by making certain plausible assumptions about individual decision-making

regarding wealth accumulation in the face of uncertainty. These axioms – comparability (or

completeness), transitivity (or consistency), strong independence, measurability and ranking –

establish the circumstances required for consistent and rational preferences (Copeland and

Weston, 1988). To these axioms is added the assumption that individuals prefer more wealth

to less (that is, they are greedy) so that the marginal utility of wealth is always positive,

although diminishing with increasing wealth. Given these circumstances, individuals will

always seek to optimise by maximising their expected utility of wealth written as follows:

E[U(W)] = ∑ Pi U(Wi) i

where Pi = probability of wealth outcome Wi

7

Individuals will use this as their objective function, and they will be seen as calculating the

expected utility of wealth for all possible alternative outcomes and then choosing the outcome

that maximises their expected utility of wealth (Copeland and Weston, 1988). According to

Von Neuman and Morgenstern’s (1944) expected utility model, rational decision-making is

based on a multiplication of the values of possible outcomes by the known objective

probabilities of their occurrence. Recognising the reality that probabilities are rarely

objectively known, Savage’s (1964) subjective expected utility model calls for multiplying

the values of outcomes by the probability of their occurrence as estimated by the individual

decision-makers.

The utility function as described above is strictly concave to the wealth axis. This

necessarily means that individuals are risk averse over all levels of wealth. In other words, a

risk premium is required in order to induce financial agents to undertake a risky alternative.

The risk premium is the difference between an individual’s expected wealth, given the risky

alternative, and the level of wealth that individual would accept with certainty if the risky

alternative were removed (his or her certainty equivalent wealth). Copeland and Weston

(1988) refer to this as the ‘Markowtiz’ risk premium, reflecting the requirement that risky

alternatives must be judged in the context of a wealth portfolio.

Bayes’ Theorem is concerned with how individuals revise their estimates of the

probabilities of some events or outcomes of interest in the light of new information that

becomes available. The individual starts with initial or prior probability estimates, receives

new information and then applying Bayes’ Theorem calculates revised or posterior

probabilities. Where there are n mutually exclusive events or outcomes A1, A2, . . . An, one of

which must occur, and B represents new information, then the posterior probability P(Ai|B)

can be calculated using Bayes’ Theorem as follows:

8

P(Ai)P(B|Ai) P(Ai|B) = __________________________________________

P(A1) P(B|A1) + P(A2) P(B|A2) + . . . P(An) P(B|An)

where i = 1, 2, . . . n

Bayesian forecasting means that probabilistic judgements are updated by the appropriate use

of Bayes’ Theorem when new information is received. Rational expectations means that

probabilistic judgements are consistent and unbiased, and they are reached with full access to

relevant new information at the time a decision is made. This does not mean that expectations

or beliefs are always accurate. Forecasting errors can occur, but they are neither biased nor

predictable (Pass et al., 1988).

As confining and unrealistic the assumption of strict rationality of financial agents is, it

has nevertheless been a powerful influence in shaping research and teaching on MFT over

many decades. Finance scholars have defended reliance upon strict rationality on a variety of

grounds that are summarised by Conlisk (1996) in the following terms:

• Since, more or less by definition, the scholarly field of finance is the study of

optimising behaviour, less than strictly rational behaviour is the concern of other

disciplines (psychology, say). Clearly, this represents a very narrow view of the

aspirations and potential of MFT.

• It is believed that, without the discipline of an optimising model, MFT would

degenerate into ad hoc prescriptions which lack overall cohesion, scientific

justification and precision. In contrast, because the strict rationality assumption is

amenable to mathematical representation, it confers tractable analysis and definite

outcomes. Thus, many financial theorists continue to assume rationality because they

think they have no alternative (Thaler, 1991).

9

• Strict rationality captures a financial agent’s best opportunity for gain. Because it is

implausible for a financial agent to forgo opportunities for gain, strict rationality

identifies the agent’s most likely actions. The necessity here is obviously that

opportunities for gain must be fully recognised.

• While real world financial agents may not actually be strictly rational, in many

circumstances they behave as if they are at the margin. This influential defence is

identified with Friedman (1953) who strongly believes that the true test of a theory is

not the validity of its underlying assumptions, but the veracity of its predictions.

• Though financial agents may not be strictly rational, they learn optimising behaviour

through repeated practice over time, and thus end up acting as if strictly rational. This

argument requires, of course, that financial agents actually learn from their mistakes

and do not simply repeat them. Furthermore, it assumes that if particular types of

decisions are made often enough, and with adequate feedback, that learning can take

place. This might not be the case for major corporate finance decisions such as

investment and financing choices. Thaler (2000, p. 135) points out that ‘Most

[financial] models have no reason to introduce learning because agents are assumed to

solve the relevant problem correctly at trial one’. At the very least, financial agents are

believed to learn from their mistakes quickly.

• Financial agents who do not behave with strict rationality may be exploited by other

financial agents who do (as arbitrageurs), and the former are not likely to survive in

the longer term. This argument presumes that there are no limits to arbitrage activities

as there might well be for corporate finance decisions. The most evident arbitrage

opportunity for bad corporate finance decisions is takeover, which has high transaction

costs and involves considerable unique risk.

After critically appraising these defences, Conlisk (1996, p. 686) concludes that:

10

. . . the standard arguments for unbounded rationality, despite their great influence, are

too extreme to be convincing. Put in more flexible form, however, the arguments

contain many useful insights about conditions favouring one or another treatment of

rationality.

Essentially, this suggests that the degree of rationality that can be attributed to financial agents

will vary with the context being studied, depending on matters such as deliberation costs,

complexity, incentives, experience, and market discipline (Conlisk, 1996; Thaler, 2000).

Bounded Rationality

An early challenge to the strict rationality and optimising behaviour of financial agents in

MFT came from Simon (1947) who introduced the possibility of bounded rationality and

‘satisficing’ behaviour. Barberis and Thaler (2003, p. 1055) indicate that one departure from

strict rationality of decision-makers:

. . . is to retain individual rationality but to relax the consistent beliefs assumption: while

investors apply Bayes’ law correctly, they lack the information required to know the

actual distribution variables are drawn from. This line of research is sometimes referred

to as the literature on bounded rationality . . .

Conlisk (1996, p. 686) argues the need for MFT to contemplate bounded rationality in the

following terms:

Human cognition is a scarce resource, implying that deliberation about [financial]

decisions is a costly activity. To avoid a free lunch fallacy, it can be argued, we are

forced to incorporate deliberation cost, and thus bounded rationality, in [financial]

models.

On the basis of his research in business organisations, Simon (1947) observes that when

faced with a problem requiring a decision managers frequently do not have the resources

(including time) to identify all possible courses of action, to evaluate each course of action

against all relevant criteria, and to choose the best alternative for implementation. While

rational, human beings only have a limited capacity to gather, store, process and understand

11

all the information required to decide an optimal response to a complex problem. They are

more likely to produce a simplified model of the problem and to sequentially review the most

obvious alternatives until they find one that is just good enough by a limited range of criteria

to address the problem, and then to cease their search. Thus, they are said to satisfice by

discovering a satisfactory and sufficient solution, rather than finding an optimal solution.

Satisficing can be seen as essentially a form of constrained optimisation reflecting the impact

of decision deliberation costs. Satisficing individuals may be procedurally rational (intending

to be rational) but are not necessarily substantively rational (achieving an optimal outcome)

(Schwartz, 1998).

Quasi-Rationality

As suggested in the introduction to the paper, a further challenge to the strict rationality and

optimising behaviour of financial agents in MFT has been provided by research in cognitive

psychology pioneered by Tversky and Kahneman (1971, 1973, 1974, 1981, 1983, 1986, 1992)

and Kahneman and Tversky (1972, 1973, 1979, 1984). This research introduces the

possibility that, in their decision-making, financial agents employ heuristics and exhibit

certain systematic cognitive biases that together lead to significant departures from the tenets

of MFT. Following the lead of Thaler (1991, 1999, 2000), the resulting circumstance will be

referred to as quasi-rationality by which Thaler (2000, p. 136) means ‘trying hard but subject

to systematic error’. Schwartz (1998, p. 46) attempts to capture the relationship between strict

rationality, bounded rationality and quasi-rationality as follows:

It may be useful to think of a continuum of rationality, with the traditional neoclassical

economist’s concept of rationality (perfect rationality) at one end and blatantly irrational

behavior at the other. Simon’s [1947] concept of bounded rationality would fall in

between, though towards the perfect rationality end. The decision making captured in

the studies of psychologists, which incorporate systematic biases, reflects what Simon

[1947] may have had in mind when he explained the concept of bounded rationality,

12

and probably some other elements as well. Thus, the decision making reflected in the

analyses of the psychologists also would fall in between the two extremes, though

further from the perfect rationality end.

This sub-section of the paper begins by briefly introducing an important theoretical

underpinning for quasi-rational finance (that is, BF). It will go on to address, in general terms,

the meaning and import of decision-making heuristics and cognitive biases. The following

section of the paper will then catalogue and explain a range of specific instances of these

phenomena.

Experimental work in the decades after Von Neuman and Morgenstern’s (1944) and

Savage’s (1964) research has revealed that individuals systematically violate expected utility

theory when choosing among risky alternatives. As a consequence, there have evolved a

plethora of non-expected utility theories, of which prospect theory developed by two

cognitive psychologists (Kahneman and Tversky, 1979; Tversky and Kahneman, 1992) is

considered by BF scholars to be the most promising for financial applications. Barberis and

Thaler (2003, p. 1069) put prospect theory into perspective as follows:

. . . prospect theory has no aspirations as a normative theory: it simply tries to capture

people’s attitudes to risky gambles as parsimoniously as possible. Indeed, Tversky and

Kahneman (1986) argue convincingly that normative approaches are doomed to failure,

because people routinely make choices that are simply impossible to justify on

normative grounds . . .

Kahneman and Tversky’s (1979) descriptive or data-driven prospect theory focuses on the

concept of subjective value – gains or losses determined with respect to a reference point.

Individuals have different reference points which can change over time. Through experiments

it has been found that subjective value is defined over gains and losses rather than over final

wealth positions. In other words, individuals do not evaluate risky alternatives in a portfolio

context.

13

It has also been found that the function relating losses to subjective value is steeper than

the function relating gains to subjective value. This means that, for a given amount, losses

tend to loom larger than gains in the minds of individuals and in their decisions. This disparity

in the value function for gains and losses leads to the prediction, confirmed by much research,

that individuals tend to be risk averse with respect to gains but risk seeking with respect to

losses – a characteristic referred to as loss aversion. Baron (2004, p. 225) provides the

following example to illustrate this finding:

. . . when asked which they prefer, a 50% chance of losing US$1,000 or a certain loss of

US$500, a large majority of persons choose the former despite the fact that the expected

values are identical in both cases; they are risk-seeking since this choice is framed in

terms of losses. However, when asked whether they prefer a certain gain of US$500 or a

50% chance of gaining US$1,000, most prefer the former; since they are focused on

gains, they are risk-averse and prefer the “sure thing”.

The finding that people appear to have a greater sensitivity to losses than to gains may explain

the often noticed preoccupation with downside risk in making business decisions. Finally, it

seems that individuals overweight small probabilities and underweight moderate and high

probabilities. The willingness to purchase lottery tickets with a very low probability of

success illustrates this point. Clearly, lottery ticket buyers believe that the likelihood of

experiencing a positive outcome is much higher than objective data suggest, a characteristic

referred to as an excessive optimism.

Overall, the chief attraction of prospect theory to BF scholars is its ability to

accommodate heuristics and cognitive biases evidenced in real world decision-making that

simply are not countenanced in expected utility theory. In general terms, the meaning and

import of decision-making heuristics and cognitive biases are now considered. It is useful to

begin by establishing that the two phenomena can be distinguished from each other.

According to Haley and Stumpf (1989, p. 481):

14

Researchers often use the terms bias and heuristic interchangeably; but important

differences distinguish these concepts. Biases occur when cognitive abilities limit

capacities; biases generally culminate in inferior decisions. Decision heuristics, may, or

may not, alter the qualities of decision outcomes. If errors result, they stem from

inaccurate premises about the data, and/or from the inference processes. Since decision

heuristics may lead to biases that affect premises and inference processes, ambiguities

distort these distinctions.

The difficulty of actually separating heuristics from cognitive biases will become apparent in

the following section of the paper.

The term heuristic is derived from the Greek word eurisco meaning ‘I discover’.

Buchanan and Huczynski (2004, p. 762) define heuristics as ‘simple and approximate rules,

guiding procedures, shortcuts or strategies that are used to solve problems’. Haley and Stumpf

(1989) draw attention to numerous heuristics, or rules-of-thumb, that individuals use to make

decisions. They indicate that heuristics influence the alternatives that decision-makers

generate, select and evaluate. Heuristics are used as filtering and organising devices, thereby

reducing the complexities of decisions and speeding up the decision process in the face of

considerable uncertainty and ambiguity. Heuristics may quickly yield acceptable solutions to

problems in an effective and efficient manner and are therefore very economical, especially in

terms of information requirements. The use of heuristics has also been associated with faster

learning and innovativeness in the sense of generating new insights into unsolved problems

and opportunities. While helpful in many situations, it must be recognised that heuristics can

lead to severe errors and systematically biased decisions. However, Conlisk (1996, p. 671)

defends the use of heuristics as follows:

Why not condemn problem solving which leads to systematic error? The answer is

simple. Deliberation cost. For a [quasi-rational] individual, heuristics often provide an

adequate solution cheaply whereas more elaborate approaches would be unduly

expensive.

15

Examples of heuristics considered in the following section of the paper are representativeness,

anchoring and adjustment, availability and framing.

Buchanan and Huczynski (2004, p. 762) define a cognitive bias as a ‘prejudiced

predisposition or a systematic distortion’ when making decisions. As subjective or

predisposed opinions, biases operate at the subconscious level, are difficult to detect, and they

have a potent and immediate impact upon an individual’s judgement. Although biases help

individuals to cope with their cognitive limitations, they may result in less rational, less

comprehensive decision-making because they systematically violate the laws of probability.

Biases may not only distort perceptions of likely outcomes, but also distort perceptions of the

risk or uncertainty associated with those outcomes. In other words, biases cause individuals to

overestimate the reliability and validity of information, to draw incorrect conclusions, and to

give information too much or too little weight. The decision circumstances in which cognitive

biases are more likely to be exhibited include (inter alia) information overload, high

uncertainty, great complexity, considerable ambiguity, high novelty, strong emotions, time

pressure and fatigue. Examples of cognitive biases considered in the following section of the

paper are overconfidence, excessive optimism, sample size neglect, illusion of control,

planning fallacy, hindsight bias, counterfactual thinking and escalation of commitment.

Because individuals are generally unaware that they exhibit biases, which are most

often applied in an unconscious manner, it is argued that it is extremely difficult to eliminate

the impact of biases upon decision-making. Barberis and Thaler (2003, p. 1068) highlight the

persistence of cognitive biases as follows:

Economists are sometimes wary of this body of experimental evidence [that is, BF]

because they believe (i) that people, through repetition, will learn their way out of

biases; (ii) that experts in a field, such as traders in an investment bank, will make

fewer errors; and (iii) that with more powerful incentives, the effects will disappear.

While all these factors can attenuate biases to some extent, there is little evidence that

16

they wipe them out altogether. The effect of learning is often muted by errors of

application: when the bias is explained, people often understand it, but then

immediately proceed to violate it again in specific applications. Expertise, too, is often

a hindrance rather than a help: experts, armed with their sophisticated models, have

been found to exhibit more overconfidence than laymen, particularly when they

receive limited feedback about their predictions. Finally, in a review of dozens of

studies on the topic, Camerer and Hogarth (1999, p. 7) conclude that while incentives

can sometimes reduce the biases people display, ‘no replicated study has made

rationality violations disappear purely be raising incentives’.

Conlisk (1996, p. 671) concurs with this view, indicating that ‘The prevailing overall

impression is that biases are not fragile effects which easily disappear, but rather substantial

and important behavioral regularities’. It would appear, therefore, that the most that can be

expected is to be able to ameliorate the negative outcomes stemming from various biases if

and when they are recognised.

As has been seen, financial models generally build on two aspects of the behaviour of

financial agents:

• How they formulate their expectations (or beliefs).

• How they ascertain and realise their preferences.

Hence the next section of the paper considers how specific heuristics and cognitive biases

impact upon these two matters.

Heuristics and Cognitive Biases

Representativeness

Decision-makers employ this heuristic when they generalise about a person or an event based

on only a few attributes of that person or only a few observations of similar events.

Representativeness relies upon being able to discern similarities between the specific

attributes of given instances and the defining attributes of classes of such instances. Thus,

17

judgements are made on the basis of how well circumstances represent or match particular

stereotypes that have emerged from past experience.

While the representativeness heuristic can be helpful, it is associated with some significant

biases:

• Base rate (or prior probability) neglect which causes individuals to behave in a manner

inconsistent with Bayes’ Theorem. In tackling their present circumstance, they

persistently ignore base rate information associated with similar situations in the past,

about which a considerable amount may be known. Instead, they tend to focus on

outcomes in the immediate past in reaching decisions, disregarding older information

such as long-run averages or statistical odds that may be relevant.

• Sample size neglect (or belief in the law of small numbers) which causes individuals

to generalise from small, non-random samples. This bias is evident when individuals

use a limited number of information inputs to draw firm conclusions about a much

larger population. The most common type of small, non-random sample used as a

basis for generalisation is, of course, the personal experience of an individual

decision-maker. Clearly, a small non-random sample may not adequately represent the

population and insensitivity to the limitations of small samples violates statistical rules

in systematic ways. Furthermore, small samples can contain a disproportionate

number of successes because failures are less likely to be well publicised and

remembered. Even if individuals gather feedback through an impartial process, the

smaller the sample, the greater the chances of receiving only positive information.

This suggests that sample size neglect may (inter alia) affect one’s perception of risk.

Employing the representativeness heuristic can impair the ability of individuals to

discriminate opportunities from background noise. A circumstance that has unique

18

characteristics may nevertheless be judged and classified by its stereotypical qualities, and an

opportunity may be lost. Thus, representativeness assures that stereotyped thinking prevails.

Anchoring and Adjustment

This heuristic may be used when a variable must be estimated in the face of considerable

uncertainty and a beginning point has to be found. As a fixed point of reference, individuals

frequently start with some initial value – an anchor value – which may or may not be

determined arbitrarily. Often the anchor will be an historical value for the variable in

question. To reach a final decision, the individual must then adjust upwards or downwards

from the anchor to take account of the particular circumstances faced that may differ from

those previously prevailing. A problem arises when, as experimental evidence suggests, the

adjustment from the anchor is insufficient to reflect the changed circumstances as revealed in

new information. In other words, an individual may anchor too much on the initial value and

his or her estimate of the variable is therefore biased towards this initial value. The anchoring

and adjustment heuristic can bias perceptions of variability and cause decision-makers to

underestimate spread. It may also inhibit innovation if used to anchor perceptions on existing

situations rather than contemplating the full range of possibilities. Overall, then, anchoring

limits the chance for additional information to produce change through the decision being

made.

Availability

Decision-makers make use of this heuristic when they estimate the likelihood of an event

occurring, and judge the appropriateness of a response to the event, by the ease with which

they can recall similar events and responses to them. While this behaviour is rational, it can

produce biased assessments because not all memories are equally retrievable or available to

recall. Events that occur more frequently, are more recent, are more salient, are more vivid,

19

are more dramatic, and are more personally relevant come to mind more easily. These

available memories will be more heavily weighted in decision-making, and are more likely to

colour perceptions of the circumstances.

Framing

This heuristic focuses attention upon the way a problem is presented to a decision-maker

and/or on how the decision-maker chooses to think about the problem (referred to as mental

accounting). In other words, regard has to be paid to how the problem is framed.

Experimental studies have found that the framing of a decision can have a profound effect on

the ultimate choice that is made. For example, it can make a significant difference to a

decision outcome if the problem is framed in terms of losses as opposed to gains. Recall that

prospect theory finds that people appear to have a greater sensitivity to losses than to gains.

Individuals tend to be risk averse with respect to gains but risk seeking with respect to losses

(referred to earlier as loss aversion). It is an important feature of prospect theory that it can

accommodate the effects of problem description or framing. MFT cannot accommodate the

effects of framing because a principle of rational decision-making is that choices should be

independent of the problem description or representation.

In MFT, broad framing is considered essential to rational decision-making. However,

research has revealed that individuals tend to frame risky decisions narrowly, possibly as a

means of dealing with complexity and/or uncertainty. Thus, they deal with one decision at a

time; with little attention being given to connections between decisions at a particular time or

over a period of time. This explains (inter alia) the finding that individuals do not evaluate

risky alternatives in a portfolio context.

20

Counterfactual Thinking

This heuristic arises from a tendency of individuals to dwell on the past, imagining what

might have been if they had made different decisions or acted differently or if the

circumstances had been different. Research has revealed that such mental simulations of

events that never occurred can have strong effects upon an individual’s emotional state, and

upon his or her learning process. Typically, an individual engaging in counterfactual thinking

focuses on imagined outcomes better than those actually obtained. This can result in intense

feelings of disappointment and regret which colour the individual’s perceptions of past

achievements and future opportunities, and which are likely to impact upon his or her future

decisions and behaviour. It appears that the nature of the regret experienced changes over

time. With respect to recent events, individuals’ tend to regret actions and decisions that

yielded disappointing results. In the longer term, however, regret tends to focus on actions

and decisions which were not undertaken and which represent missed opportunities. Aversion

to regret – the pain felt when it is found out, too late, that different choices would have led to

better results – is frequently recognised as a separate bias.

Counterfactual thinking can enhance an individual’s experiential learning by

improving understanding of the likely causes of particular events or outcomes. By imagining

events and outcomes that did not occur, an individual often gains insights into the factors that

resulted in the events and outcomes that were actually experienced. Such insights may

contribute to improved performance by suggesting better strategies, increasing the expectation

of positive results, and increasing feelings of personal control. Overall, then, evidence

suggests that counterfactual thinking can produce both benefits and costs, with the possibility

that the benefits in terms of learning may exceed the costs.

21

Herding

This heuristic is followed when individuals seek safety in numbers by following financial

trends, fads or fashions established by others, rather than making their own independent

judgements. Herding occurs for a number of reasons. First, individuals like to believe they are

prudent. Thus, if a decision is made that has a negative outcome, the individual will mind less

about the loss if he or she thinks other sensible individuals would have made the same

decision. Second, individuals give too much emphasis to recent data, and not enough to base

rates (or prior probabilities). Hence, they tend to focus on very recent trends when reaching

decisions. Third, the cost and difficulty of gathering and processing information means that

observing the choices of others is often a cheap and satisfactory alternative. The chief

problem with herding is, of course, that a consensus view is not necessarily correct – all may

be subject to the same error(s).

Reasoning by Analogy

This heuristic is often employed when individuals are faced with novel and complex

circumstances. Reasoning by analogy involves the application of simple analogies and images

to guide problem definition and sensemaking. This process may help to reduce uncertainty,

and might even yield creative solutions. However, it may also provide extremely simplistic

representations of intrinsically complex situations. Individuals may mistakenly believe that

their problems are simpler or more familiar than they really are, or the analogy used may not

be an appropriate match to the presenting situation.

Overconfidence

This widely observed bias arises when individuals tend to overestimate the correctness of

their initial estimates when answering moderate to difficult questions or when dealing with

ill-structured decision situations. Because of their overconfidence, they do not revise their

22

initial estimates even after receiving new information. From a statistical viewpoint,

overconfidence is evident when the confidence intervals individuals assign to their estimates

are far too narrow and/or they are poorly calibrated when estimating probabilities so that

events they think are certain occur less frequently than they should and events they believe are

impossible happen more frequently than they should. Individuals exhibiting overconfidence

tend to treat their assumptions as facts and do not see uncertainty associated with conclusions

stemming from those assumptions. They may therefore erroneously conclude that a certain

action or decision is not risky.

It has been argued that overconfidence arises from a lack of meta-knowledge, meaning

that individuals are unaware of the limits of their knowledge and therefore they are

overconfident when making forecasts. In other words, individuals are claimed to not know

what they don’t know. The overconfidence bias is also variously attributed to the anchoring

and adjustment and the availability heuristics and to self-attribution, hindsight and

confirmation biases (see below). Research has shown that some classes of individuals exhibit

higher levels of overconfidence than others. Moreover, it appears that individuals are more

confident of their predictions in fields where they have self-declared expertise.

Excessive Optimism

This pervasive bias is exhibited when an individual systematically overestimates the

probability of a favourable outcome and/or systematically underestimates the probability of an

unfavourable outcome. Considerable empirical evidence suggests that most people have

unrealistically rosy views of their abilities and prospects, and they are therefore excessively

optimistic about future events. Several factors contribute to this phenomenon: positive self-

evaluations, high personal commitment, and a strong sense of control. Thus, excessive

optimism is associated with the tendency of individuals to have a high personal regard for

23

their own abilities or competence, regardless of objective evidence to the contrary. Individuals

are more optimistic about outcomes when they are fervently committed to achieving those

outcomes because their wealth, reputation, employability, etc. would otherwise suffer.

Finally, greater optimism can result when individuals believe, rightly or wrongly, that they

can exercise effective control over their activities and plans, thus diminishing the perceived

risk of failure (see illusion of control below).

Excessive optimism may have both beneficial and harmful consequences. On the

beneficial side, it may aid in maintaining a relatively high level of self-esteem. Furthermore,

the illusion of invulnerability resulting from excessive optimism may reduce anxiety and

enable individuals to function without being overcome by trepidation or fear. On the harmful

side, excessive optimism may cause perceptions of risk to be lower, thus discouraging an

individual from taking precautions to avoid adverse outcomes. Research has shown that

excessive optimism is quite resistant to de-biasing interventions such as awareness raising and

advice.

Illusion of Control

This bias arises when an individual overemphasises the extent to which his or her skill can

increase performance in situations where chance plays a large part and skill is not necessarily

the deciding factor. Individuals exhibiting illusion of control have a greater expectancy of

success than objective probability would suggest because they believe their skills are more

highly developed than those of others. Thus, an illusion of control can contribute to an

excessive optimism bias.

There are two main reasons for the illusion of control bias. First, individuals are

motivated to control their environment and they derive personal satisfaction from belief in

their own competence in accurately predicting and controlling the outcome of uncertain future

24

events. Second, it is frequently very difficult to ascertain whether a particular outcome is a

consequence of exercising skill or simply a chance occurrence, or a combination of both.

Individuals exhibiting an illusion of control do not respond differently to controllable and

uncontrollable circumstances. The illusion of control reduces the anxiety experienced in the

face of uncertainty and may cause the individual to underestimate the level of risk faced

because they believe their skills can prevent negative outcomes.

Planning Fallacy

This bias refers to the general tendency of individuals to overestimate the amount that they

can achieve in a specific time; or alternatively, underestimate the amount of time that will be

necessary to complete a specific task. The planning fallacy arises because individuals tend to

ignore past situations and experiences with similar characteristics when making predictions

about future outcomes. They are also inclined to treat the current situation as if it is unique

and full of uncertainties, thus rendering past situations and experiences irrelevant. If they have

been late in completing tasks in the past, they often blame it upon external factors beyond

their control – an example of self-serving bias (see below).

Self-Attribution or Self-Serving Bias

This bias alludes to the strong tendency of most individuals to attribute success to internal

causes – for example, to their own skill, sound judgement or hard work; and to attribute

unsuccessful outcomes to external causes such as factors beyond their control, the actions of

others or bad luck. Succumbing to this bias frequently can lead individuals to the attractive if

erroneous belief that they are very capable.

Hindsight Bias

This bias arises from the tendency of individuals to believe, after an outcome has occurred,

that they had been able to foresee it happening. Past events are seen as more predictable than

25

they actually were. If individuals believe that they anticipated the past better than they

actually did, they may also believe that they can predict the future better than they really can.

Furthermore, hindsight bias may make it more difficult for individuals to admit their

mistakes. After all, if they could have forecast negative outcomes, why did they not do

something to avoid them? This bias may hinder learning from experience.

Escalation of Commitment

This bias refers to the tendency, under certain conditions, for an individual who has made an

initial decision to become overly committed to the original choice despite negative feedback;

and to make further decisions that are biased by this commitment. Thus, escalation of

commitment results in a determination to further pursue a course of action when the available

evidence suggests that this is not appropriate. There are several factors that heighten

escalation of commitment. First, research suggests that the more negative the feedback the

greater may be the commitment. It appears that an individual in these circumstances is prone

to engage in self-justification. Second, the more responsible for the initial decision an

individual feels the more likely it is he or she will view reversing the decision as backing

away from such responsibility. Third, escalation of commitment may be greater when the

decision-maker is overconfident. Fourth, the more cognitive effort and skill the initial

decision entailed the more reluctant might the decision-maker be to begin the process over

again. Fifth, the greater the visibility of the initial decision to external parties the more likely

the individual is to attempt to avoid loss of face by admitting failure to others. It appears that

learning ultimately does take place and escalation of commitment may disappear after several

trials and prolonged negative feedback.

26

Confirmation Bias or Belief Perseverance

This bias alludes to the tendency for information that confirms an individual’s current beliefs

(or, at least, is consistent with them) to be noticed, processed and remembered more readily

than information that disconfirms current beliefs. Thus, positive, confirming evidence is

weighed more heavily than negative, disconfirming evidence with respect to given

alternatives. The counsel of others with contrary views may be disregarded or treated with

excessive scepticism. Individuals may be reluctant to search for evidence that contradicts their

beliefs or might even misinterpret evidence that goes against their position as actually being

in its favour. Cognitive dissonance which involves fitting beliefs to convenience may be

evident. When there is a conflict between an individual’s beliefs and reality, he or she may try

to rationalise irrational behaviour. Confirmation bias is damaging to decision-making because

it prevents the true state of a situation from being known and it reinforces preconceptions and

prejudices.

Sunk Cost Fallacy

This most common of decision-making biases occurs when individuals allow their choices

between future alternatives to be influenced by costs incurred at some time in the past which

will be unchanged. MFT holds that only incremental costs and benefits should affect

decisions about future events, and that taking account of sunk costs is irrational. It appears

that sunk costs increase an individual’s aspiration level – the outcome anticipated in

accordance with inputs. Those who have invested in sunk costs perceive outcomes below the

aspiration level as being more negative. Sunk costs also cause individuals to be more risk

seeking than they would have been if they had not incurred these costs.

27

Endowment Effect

This bias reflects the propensity of individuals to value what they have more highly than they

would an opportunity to newly acquire the same good. Assuming no information advantage,

MFT holds that ownership of an asset should not affect its valuation. However, prospect

theory provides an explanation for the endowment effect in terms of loss aversion. When a

person owns an object its loss has greater magnitude than the value exchanged when the

identical object is gained from the market-place. The endowment effect has been found to

increase with the duration of ownership.

The BF literature suggests that the relationship between ownership and value is

moderated by the amount of sunk costs and how the object was obtained. It appears that

attachment to an object depends on whether it was obtained by one’s own efforts or by a

chance outcome – a phenomenon referred to as source dependence. Unearned or windfall

gains are not seen to be as valuable as earned gains, and are therefore more readily spent or

gambled.

Ambiguity Aversion

This bias is exhibited when individuals avoid decision situations in which they are uncertain

about the probability distribution of outcomes. Such circumstances are known as situations of

ambiguity, and the dislike of them as ambiguity aversion. Subjective expected utility theory

does not allow financial agents to express their level of confidence regarding a probability

distribution, and so cannot capture ambiguity aversion. Evidence suggests that the level of

ambiguity aversion depends on how competent an individual feels in assessing the relevant

distribution. Someone who, from past experience, is familiar with a particular risky situation

may feel more able to judge the probability of outcomes and therefore display less ambiguity

aversion.

28

Conservatism or Status Quo Bias

While representativeness leads to underweighting of base rates (or prior probabilities), there

are situations where base rates are overweighted relative to more recent evidence. This

overweighting causes individuals to have a conservative bias in their decision-making which

leads to a preference for the status quo over significant change.

Social Desirability Bias

Placing importance upon interpersonal relations and the approval of others, decision-makers

succumb to this bias when they believe that solutions become more effective as more

individuals support the solutions. Decision-makers exhibit social desirability bias when they

do what they think other people want them to do, rather than what they actually feel they

should do. Acute need for acceptance by others prompts them to promote the ideas of others

instead of their own, almost without regard to their relative merits.

Discussion and Conclusions

To even the most naïve observer of human nature, the outright plausibility of the heuristics

and cognitive biases described make it difficult to dismiss BF as a passing fad promoted by

financial heretics. When the rapidly growing empirical evidence on the existence and

operation of these heuristics and cognitive biases is also taken into account, it is clear that

some way of accommodating the tenets of BF must be found. The emerging consensus seems

to be that, rather than rejecting MFT in favour of an alternate world view, MFT needs to be

extended to capture the complexities of real world financial phenomena revealed by BF. For

example, Conlisk (1996, p. 672) expresses the following opinion:

In summary, the bias evidence suggests that people are capable of a wide variety of

substantial and systematic reasoning errors relevant to [financial] decisions. Further, the

evidence suggests that the magnitude and nature of the errors are themselves

29

systematically related to economic conditions such as deliberation cost, incentives, and

experience. In this sense, investigation of [quasi-rationality] is not a departure from

[financial] reasoning, but a needed extension of it.

Thaler (1999, p. 16) takes this line of argument to its logical conclusion as follows:

I predict that in the not-too-distant future, the term “behavioral finance” will be

correctly viewed as a redundant phrase. What other kind of finance is there? In their

enlightenment, [finance scholars] will routinely incorporate as much “behavior” into

their models as they observe in the real world. After all, to do otherwise would be

irrational.

In this sense then, the normative perspective of MFT is likely to blend with the descriptive

perspective of BF to give a more complete picture of how financial agents can and do meet

their objectives through quasi-rational decision-making.

As far as future research is concerned there would appear to be a real opportunity for re-

examining various aspects of corporate finance that may be illuminated in a BF framework. In

a review article on BF, Thaler (1999, p. 16) indicates ‘I would like to see more behavioral

finance research in the field of corporate finance. Most of the research so far has been in the

field of asset pricing; much less has been done on corporate finance’. Thaler (1999) cites

Stein’s (1996) work on capital budgeting in an irrational world as an example of the type of

research needed. Heaton (2002, p.33) has more recently indicated that ‘little work in corporate

finance has dropped the assumption that managers are fully rational’. He identifies Roll

(1986) on managerial overconfidence and takeovers, DeMeza and Southey (1996) on

managerial optimism and entrepreneurship, and Boehmer and Netter (1997) on managerial

optimism and corporate acquisitions as notable exceptions. Heaton (2002) himself has

conducted research on managerial optimism and corporate finance. Other examples of the

application of BF to the study of corporate finance are Shleifer and Vishny (1990) on

corporate investments; Shefrin and Statman (1984), Frankfurter and McGoun (2000), and

30

Frankfurter et al. (2002) on corporate dividends; Ritter (1991), Loughran (2002), Owen

(2002), Burton et al. (2003), and Mohan and Chen (2004) on initial public offerings; and

Sayrak and Shukla (2005) on corporate governance. Thus, while a start has been made, there

would seem to be ample opportunities for re-examining the enduring puzzles of corporate

finance through the lens of BF. Capital structure decisions and dividend policy are clearly

priorities for this effort.

References

Barberis, N. and Thaler, R., 2003, ‘A survey of behavioral finance’, in G.M. Constantinides,

M. Harris and R. Stulz eds Handbook of the Economics of Finance, Elsevier Science B.V.,

Amsterdam, pp. 1053-1123.

Baron, R.A., 2004, ‘The cognitive perspective: a valuable tool for answering

entrepreneurship’s basic “why” questions’, Journal of Business Venturing, vol. 19, no. 2,

pp. 221-239.

Black, F., 1976, ‘The dividend puzzle’, Journal of Portfolio Management, vol. 2, no. 2, pp. 5-

8.

Black, F. and Scholes, M., 1973, ‘The pricing of options and corporate liabilities’, Journal of

Political Economy, vol. 81, no. 3, pp. 637-654.

Boehmer, E. and Netter, J.M., 1997, ‘Management optimism and corporate acquisitions:

evidence from insider trading’, Managerial and Decision Economics, vol. 18, no. 7/8, pp.

693-708.

Buchanan, D. and Huczynski, A., 2004, Organizational Behaviour: An Introductory Text, 5th

Edition, Prentice Hall, Harlow, England.

31

Burton, B., Helliar, C. and Power, D., 2003, ‘A behavioural perspective on IPOs and SEOs’,

Research Report No. 82, Association of Chartered Certified Accountants, London,

England.

Camerer, C. and Hogarth, R., 1999, ‘The effects of financial incentives in experiments: a

review and capital-labor production framework’, Journal of Risk and Uncertainty, vol. 19,

no. 1, pp. 7-42.

Clark, J.M., 1918, ‘Economics and modern psychology’, Journal of Political Economy, vol.

26, no. 1, pp. 1-30.

Conlisk, J., 1996, ‘Why bounded rationality’, Journal of Economic Literature, vol. 34, no. 2,

pp. 669-700.

Copeland, T.E. and Weston, J.F., 1988, Financial Theory and Corporate Policy, 3rd edn,

Addison-Wesley, Reading, Massachusetts.

Dean, J., 1951, Capital Budgeting: Top-Management Policy on Plant, Equipment, and

Product Development, Columbia University Press, New York.

De Bondt, W.F.M. and Thaler, R.H., 1985, ‘Does the stock market overreact?’, Journal of

Finance, vol. 40, no. 3, pp. 793-808.

De Bondt, W.F.M. and Thaler, R.H., 1995, ‘Financial decision-making in markets and firms:

a behavioral perspective’, in R.A. Jarrow, V. Maksimovic, W.T. Ziemba, G.L. Nemhauser

and A.H.G. Rinnoy Kan eds Handbooks in Operational Research and Management

Science: Finance, Elsevier, Amsterdam, pp. 385-410.

DeMeza, D. and Southey, C., 1996, ‘The borrower’s curse: optimism, finance, and

entrepreneurship’, The Economic Journal, vol. 106, no. 435, pp. 375-386.

Fama, E.F., 1970, ‘Efficient capital markets: a review of theory and empirical work’, Journal

of Finance, vol. 25, no. 2, pp. 383-417.

32

Frankfurter, G.M. and McGoun, E.G., 2000, ‘Thought contagion and financial economics: the

dividend puzzle as a case study’, Journal of Behavioral Finance, vol. 1, no. 2, pp. 145-153.

Frankfurter, G., Schmidt, H. and Topalov, M., 2002, ‘The perception of dividends by

management’, Journal of Behavioral Finance, vol. 3, no. 4, pp. 202-217.

Friedman, M., 1953, Essays in Positive Economics, University of Chicago Press, Chicago,

Illinois.

Haley, U.C. and Stumpf, S.A., 1989, ‘Cognitive trails in strategic decision-making: linking

theories of personalities and cognitions’, Journal of Management Studies, vol. 26, no. 5,

pp. 477-497.

Heaton, J.B., 2002, ‘Managerial optimism and corporate finance’, Financial Management,

vol. 31, no. 2, pp. 33-45.

Hirshleifer, D., 2001, ‘Investor psychology and asset pricing’, Journal of Finance, vol. 56, no.

4, pp. 1533-1597.

Jensen, M.C. and Meckling, W.H., 1976, ‘Theory of the firm: managerial behavior, agency

costs and ownership structure’, Journal of Financial Economics, vol. 3, no. 4, pp. 305-360.

Kahneman, D. and Tversky, A., 1972, ‘Subjective probability: a judgement of

representativeness’, Cognitive Psychology, vol. 3, no. 3, pp. 430-454.

Kahneman, D. and Tversky, A., 1973, ‘On the psychology of prediction’, Psychological

Review, vol. 80, no. 4, pp. 237-251.

Kahneman, D. and Tversky, A., 1979, ‘Prospect theory: an analysis of decision under risk’,

Econometrica, vol. 47, no. 2, pp. 263-292.

Kahneman, D. and Tversky, A., 1984, ‘Choices, values and frames’, American Psychologist,

vol. 39, no. 4, pp. 341-350.

Leland, H.E. and Pyle, D.H., 1977, ‘Informational asymmetries, financial structure, and

financial intermediation’, Journal of Finance, vol. 32, no. 2, pp. 371-387.

33

Lintner, J., 1965a, ‘The valuation of risk assets, and the selection of risky investments in

stock portfolios and capital budgets’, Review of Economics and Statistics, vol. 45, no. 1,

pp. 13-37.

Lintner, J., 1965b, ‘Security prices, risk and maximal gains from diversification’, Journal of

Finance, vol. 20, no. 4, pp. 587-616.

Loughran, T., ‘The changing nature of the initial public offering market’, Journal of

Behavioral Finance, vol. 3, no. 1, pp. 23-28.

Markowitz, H.M., 1952, ‘Portfolio selection’, Journal of Finance, vol. 7, no. 1, pp. 77-91.

Markowitz, H.M., 1959, Portfolio Selection: Efficient Diversification of Investments, Wiley,

New York.

Miller, M.H., 1986, ‘Behavioral rationality in finance: the case of dividends’, Journal of

Business, vol. 59, no. 4, pp. s451-s468.

Miller, M.H and Modigliani, F., 1961, ‘Dividend policy, growth, and the valuation of shares’,

Journal of Business, vol. 34, no. 4, pp. 411-433.

Modigliani, F. and Miller, M.H., 1958, ‘The cost of capital, corporation finance and the

theory of investment’, American Economic Review, vol. 48, no. 3, pp. 261-297.

Modigliani, F. and Miller, M.H., 1963, ‘Corporate income taxes and the cost of capital: a

correction’, American Economic Review, vol. 53, no. 3, pp. 433-443.

Mohan, N. and Chen, C., 2004, ‘Are IPOs priced differently based upon gender?’, Journal of

Behavioral Finance, vol. 5, no. 1, pp. 57-65.

Myers, S.C., 1984, ‘The capital structure puzzle’, Journal of Finance, vol. 39, no. 3, pp. 575-

592,

Owen, S., 2000, ‘Behavioural finance and the decision to invest in high tech stocks’, Working

Paper No. 119, School of Finance and Economics, University of Technology, Sydney,

New South Wales.

34

Pass, C., Lowes, B. and Davies, L., 1988, Dictionary of Economics, Collins, London.

Ritter, J.R., 1991, ‘The long-run performance of initial public offerings’, Journal of Finance,

vol. 46, no. 1, pp. 3-27.

Roll, R., 1986, ‘The hubris hypothesis of corporate takeovers’, Journal of Business, vol. 59,

no. 2, pp. 197-216.

Ross, S.A., 1976, ‘The arbitrage theory of capital asset pricing’, Journal of Economic Theory,

vol. 13, no. 3, pp. 341-360.

Savage, L., 1964, The Foundations of Statistics, Wiley, New York.

Sayrak, A. and Shukla, L., 2005, ‘Corporate governance: incentive, conflict of interest and

bias’, Journal of Behavioral Finance, vol. 6, no. 1, pp. 2-5.

Schleifer, A., 1999, Inefficient Markets: An Introduction to Behavioural Finance, Oxford

University Press, Oxford, United Kingdom.

Schwartz, H., 1998, Rationality Gone Awry? Decision Making Inconsistent With Economic

and Financial Theory, Praeger, Westport, Connecticut.

Sharpe, W.F., 1964, ‘Capital asset prices: a theory of market equilibrium under conditions of

risk’, Journal of Finance, vol. 19, no. 3, pp. 425-442.

Shefrin, H., 1999, Beyond Greed and Fear: Understanding Behavioral Finance and the

Psychology of Investing, Harvard Business School Press, Boston, Massachusetts.

Shefrin, H.M. and Statman, M., 1984, ‘Explaining investor preference for cash dividends’,

Journal of Financial Economics, vol. 13, no. 2, pp. 253-282.

Shleifer, A. and Vishny, R.W., 1990, ‘Equilibrium short horizons of investors and firms’,

American Economic Review, vol. 80, no. 2, pp. 148-153.

Simon, H.A., 1947, Administrative Behavior, Macmillan, New York.

Statman, M., 1999, Behavioral finance: past battles and future engagements’, Financial

Analysts Journal, vol. 55, no. 6, pp. 18-27.

35

Stein, J., 1996, ‘Rational capital budgeting in an irrational world’, Journal of Business, vol.

69, no. 4, pp. 429-456.

Thaler, R.H., 1991, Quasi Rational Economics, Russel Sage Foundation, New York.

Thaler, R.H., 1992, The Winner’s Curse: Paradoxes and Anomalies of Economic Life, Russel

Sage Foundation, New York.

Thaler, R.H. ed., 1993, Advances in Behavioral Finance, Russell Sage Foundation, New

York.

Thaler, R.H., 1999, ‘The end of behavioral finance’, Financial Analysts Journal, vol. 55, no.

6, pp. 12-17.

Thaler, R.H., 2000, ‘From Homo Economicus to Homo Sapiens’, Journal of Economic

Perspectives, vol. 14, no. 1, pp. 133-141.

Tversky, A. and Kahneman, D., 1971, ‘Belief in the law of numbers’, Psychological Bulletin,

vol. 76, no. 2, pp. 105-110.

Tversky, A. and Kahneman, D., 1973, ‘Availability: a heuristic for judging frequency and

probability’, Cognitive Psychology, vol. 5, no. 2, pp. 207-232.

Tversky, A. and Kahneman, D., 1974, ‘Judgement under uncertainty: heuristics and biases’,

Science, vol. 185, no. 4157, pp. 1124-1131.

Tversky, A. and Kahneman, D., 1981, ‘The framing of decisions and psychology of choice’,

Science, vol. 211, no. 4481, pp. 453-458.

Tversky, A. and Kahneman, D., 1983, ‘Extensional versus intuitive reasoning: the

conjunction fallacy in probability judgement’, Psychological Review, vol. 90, no. 4, pp.

293-315.

Tversky, A. and Kahneman, D., 1986, ‘Rational choice and the framing of decisions’, Journal

of Business, vol. 59, no. 4, pp. 251-278.

36

Tversky, A. and Kahneman, D., 1992, ‘Advances in prospect theory: cumulative

representation of uncertainty’, Journal of Risk and Uncertainty, vol. 5, no. 4, pp. 297-323.

Von Neumann, J. and Morgenstern, O., 1944, Theory of Games and Economic Behavior,

Princeton University Press, Princeton, New Jersey.