3.

0 DECISION THEORY

Decision making can be defined as selecting a course of action from a range of alternatives OR the process
by which a course of action is chosen from available alternatives for the purpose of achieving desired
results. An outstanding quality of a successful manager is his ability to make sound and logical decisions.

Decision theory—1. is body of methods and related analytical techniques that help a decision maker
choose among a set of choices under possible consequences. Decision theory practices can help marketers
to make even better decisions based on all different probable outcomes which can aid in achieving even
higher revenues in their programs.

2. includes mathematical techniques that can help you decide among alternatives whose results would be
seen in the future.

3. is the study of a person or agents’ choices. The theory helps us understand the logic behind the choices
professionals, consumers, or even voters make. There are two branches of decision theory;

- Normative Decision Theory and

- Optimal Decision Theory.

Normative Decision Theory; Analyzes the outcomes of decisions whereas Optimal Decision Theory deals
with, the investigation and analysis of why individuals and agents of choice make the decisions that they
do.

3.2 Explanation of terms

The decision maker- refers to an individual or a group of individuals responsible for making a choice of
an appropriate course of action amongst the available course of action

State of nature- Consequences (or events) of any course of action and are dependent upon certain factors
beyond the control of the decision maker.

Course of action and state of nature: Course of action is the decision taken by the decisionmakers. A
state of nature is a future condition, which is not under the control of the decisionmaker; and it may be
linked to various courses of action. It can be probabilistic or conditional, such as a state of the economy,
a weather condition, a political development, etc. The states of nature are usually not determined by the
action of an individual or an organization. These are the result of an ‘act of God’ or the result of many
situations pushing in various directions.

Pay-off: An outcome (numerical value) resulting from each possible combination of alternatives and
states of nature is called pay-off. The pay-off values are always conditional values because of unknown
states of nature. Pay-off is measured within a specified period, which is called the decision horizon. Pay-
off can also be measured in terms of money, market share, or other measures. Pay-offs considered in most
decisions are monetary.

3.3 Decision making environments

Decision analysis is used to determine optimum strategies where a decision maker is faced with several
decisions alternatives. We may come across several decision-making situations.

a) Decision under certainty

b) Decisions under risk

c) Decisions under uncertainty

d) Decisions under conflict

a) Decision under certainty

In this case, the decision-maker has the perfect information of the alternatives and consequences of every
decision choice with certainty. Obviously, an alternative should be selected that yields the largest return
(pay-off) for the known future (state of nature). For example, the decision to either invest in public
provident fund or invest in a life insurance policy is the one in which investors have complete information
about the future (matured amount) because payment would be made when it is due. Few decisions are that
certain, and most contain risk and /or uncertainty.

b) Decision under risk

In this case, the decision-maker has less than complete knowledge of the consequence of every decision
choice because it is not definitely known which outcome will occur. This means there is more than one
state of nature and he makes an assumption of the probability with which each state of nature will occur.

Decision-making under risk is a probabilistic decision situation in which more than one state of nature
exists and the decision-maker has sufficient information to assign probability values to the likely
occurrence of each of these states. Knowing the probability distribution of the states of nature, the best
decision is to select that course of action which has the largest expected pay-off value.

Risk refers to situations where the decision maker can estimate the likelihood of the alternative outcomes,
possibly using statistical methods. Banks have developed tools to assess credit risk, and so reduce the risk
that the borrower will not repay the loan. The questions on an application form for a loan (home ownership,
duration, address, employer’s name, etc.) enable the bank to assess the risk of lending money to that
person.

c) Decisions under uncertainty

Uncertainty means that people know what they wish to achieve/goals, but do not have enough information
about alternatives and future events to estimate the risk confidently. i.e. available alternatives, the
likelihood of their occurrence and the outcomes are all unknown. Decisions made under uncertainty are
the most difficult to take because of this lack of concrete knowledge. Such decisions tend to be ambiguous,
intangible and highly unusual. In the current business environment more and more decisions are taken
under uncertainty. When making decisions under uncertain conditions, managers require intuition and
judgement.
Factors that may affect the outcomes of deciding to launch a new product (future growth in the market,
changes in customer interests, competitors’ actions) are impossible to predict. Entrepreneurs live with
very high levels of uncertainty: they cannot know if their venture will succeed, nor is it in their nature to
allow that to deter them.

In decision under uncertainity, the decision-maker cannot specify probabilities with which the various
states of nature (futures) will occur. However, this is not the case of decision-making under ignorance
because possible states of nature are known. For example, the probability that the share of a company will
grow 10 times after 5 years is not known. Thus, if the decision-maker does not know with certainty which
state of nature will occur, then he/she is said to be making a decision under uncertainty. The five
commonly used criteria for decision-making under uncertainty are as follows:

(i) Criterion of optimism (Maximax or Minimin)

(ii) Criterion of pessimism (Maximin or Minimax)

(iii) Laplace criterion

(iv) Savage (minimax regret)

(v) Hurwicz (equal likely/ Bayes’ criterion)

(i) Criterion of optimism (Maximax or minimin)

In this approach, a decision-maker has an optimistic view about the output. In maximax criterion, the
decision maker selects that particular strategy which corresponds to the maximum pay off for each
strategy.

Thus, the maximax criterion consists of the following two steps

Step I Determine the maximum possible pay-off for each alternative.

Step II Select that alternative which corresponds to the maximum of the above maximum pay-offs

In the decision problems dealing with costs, the minimum for each alternative is considered and then the
alternative which minimizes the above minimum cost is selected. This is termed as minimin principle.

(ii) Criterion of pessimism (Maximin or Minimax) /Wald criterion

This criterion is based on the conservative approach to assume that, the worst possible is going to happen.
The decision maker considers each strategy and locates the minimum pay-off for each and then selects
that alternative which maximise the minimum pay-off. Therefore, it is also known as maximin approach.
Thus, this criterion involves two steps.

Step I Find the minimum assured pay-off for each alternative (course of action/strategy/decision)

Step II Choose that alternative which corresponds to the maximum of the above minimum pay-off
When dealing with the cost, the maximum cost associated with each other alternative is considered and
the alternative that minimise this maximum cost is chosen. This is known as minimax criterion and it
involves two steps.

Step I Determine the maximum possible cost for each alternative.

Step II Choose that alternative which corresponds to the minimum of the above costs.

(iii) Laplace Criterion (equally likely decision criterion/Baye’s criterion)

This criterion is based on, what is known as the principle of insufficient reason. Since the probabilities
associated with the occurrence of various events are unknown, there is not enough information to conclude
that these probabilities will be different. Hence it is assumed that all states of nature will occur with equal
probability. That is, each state of nature is assigned an equal probability. As states of nature of mutually
exclusive and collectively exhaustive, the probability of each of these must be 1/ (number of states of
nature). In conclusion, Laplace criterion, finds the decision alternative with the highest average pay-off.
By first, calculating the average pay-off for every alternative, then selecting the alternative with maximum
average pay-off.

Laplace criterion involves following steps.

Step I Assign equal probabilities 1/(number of states of nature) to each pay off a strategy.

Step II Determine the expected pay off value for each alternative

Step III Select that alternative which corresponds to the maximum (and minimum for cost) of the
above expected pay offs.

(iv) Hurwicz criterion (Criterion of Realism)

This criterion suggests that a rational decision maker should neither be completely optimistic nor be
pessimistic and therefore, must display a mixture of both. Hurwicz, who suggests this criterion, introduced
the idea of a coefficient of optimism (denoted by α) to measure the decision maker’s degree of optimism.
This coefficient lies between 0 and 1, where 0 represents a completely pessimistic attitude about the future
and 1, completely optimistic attitude about the future. Thus, if α is the coefficient of optimism, then (1- α)
will represent the coefficient of pessimism. Under this approach, the resultant pay-off for that course of
action is calculated as sum of product of the coefficient of realism and pay-off using the following
equation:

Resultant Pay-off = α × (Maximum pay-off) + (1− α) × (Minimum pay-off)

Table 3.1 shows the format of the pay-off table.

 Table 3.1: Pay-off table

Table 3.1 shows the format of pay-off table. For the course of action S1, there are two states of nature: N1
and N2 and the corresponding pay-offs are P11 and P12. Similarly, for the course of action S2, there are
two states of nature: N1 and N2 and the corresponding pay-offs are P21 and P22.

The working procedure is –

• Decide the coefficient of optimism α and then the coefficient of pessimism (1- α)

• Determine the maximum as well as minimum pay off for each, alternative and obtain the quantities
h = α (maximum of each alternative) + (1- α) minimum of each alternative

• Select an alternative with value of h as maximum.

A difficulty with this criterion is the appropriate selection of α between 0 and 1

(v) Savage Criterion (criterion of regret, minimax regret criterion, opportunity loss decision
criterion)

The savage criterion is based on the concept of regret (or opportunity loss) and calls for selecting the
course of action that minimises the maximum regret. In this criterion, it is assumed that decision maker
feels regret after adopting a wrong course of action (alternative) resulting in an opportunity loss of pay
off. Opportunity loss (regret) equals to the difference between each pay-off and the largest pay-off for
each state of nature.

In this approach, an opportunity loss (regret) table is prepared. Then, using this regret table, the maximum
regret for each possible decision is listed. The decision chosen is the one corresponding to the minimum
of the maximum regrets

Step I From the given pay off matrix, develop an opportunity loss (or regret) matrix

- the best pay-off corresponding to each state of nature (maximum for profit and minimum
for cost)

- ith regret = (maximum pay off – ith pay off) for jth event if the pay offs represent profits

= (minimum pay off – ith pay off) for jth event if the pay offs represent costs

Step II Determine the maximum regret (opportunity loss) amount for each alternative.

Step III Choose that alternative which corresponds to the minimum regrets

d) Decision under conflict/ambiguity

Ambiguity is when people are uncertain about their goals and how best to achieve them. It is a situation
in which the intended goals are unclear, and so the alternative ways of reaching them are equally fluid –
leading to stress.
 Figure 3.1 Decision making conditions

Worked example 1

A company XYZ is thinking about the three decision alternatives: introduction of a new product by
replacing the existing product at a much higher price (S1) or effecting a moderate change in the
composition of the existing product at a small increase in price (S2) or bringing a minor change in the
composition of the existing product with a negligible increase in price (S3). The three possible states of
nature or events are: high increase in sales (N1), no change in sales (N2) and decrease in sales (N3). The
marketing department of the company has worked out the pay-offs in terms of yearly net profits for the
strategies of each of the three events (expected sales). The profits (pay-offs) for different courses of action
under various states of nature are shown in Table 3.2.

Table 3.2 Pay-off matrix

Which strategy should the company choose on the basis of;

(a) optimistic approach, (b) conservative approach, (c) minimax regret approach, (d) equally likely
(Laplace) criterion and (e) Hurwicz criterion (use α = 0.3)?

Solution

(a) Optimistic approach (Table 3.3)

Table 3.3: Pay-off table

The maximum among maximum pay-offs is Rs 750,000. Thus strategy S1 should be chosen.

(b) Conservative or pessimistic approach (Table 3.4)

Table 3.4: Pay-off table

The maximum among minimum pay-offs is Rs 350,000. Thus, strategy S3 should be chosen.
(c) Minimax regret approach (Table 3.5)

Table 3.5: Opportunity cost table

The minimum among maximum opportunity costs is Rs150,000. Thus, strategy S1 should be chosen.

(d) Equally likely (Laplace) criterion (Table 3.6)

Table 3.6: Pay-off table

Equally likely probability is 1/3 because there are three states of nature and probability of each is same.
The maximum among average values is Rs 433,333. Thus strategy S1 should be chosen.

(e) Hurwicz criterion (use α = 0.3) (Table 3.7)

 Table 3.7: Pay-off table

The maximum pay-off is Rs 365,000. Thus, strategy S1 should be selected.

3.4 The decision-making process

Most models of decision making include six essential steps that it is recommended managers should follow
when making decisions. As shown in Figure 3.2, managers should: 1. Identify and diagnose the problem.
2. Identify alternative solutions. 3. Evaluate alternatives. 4. Choose an alternative. 5. Implement the
decision. 6. Evaluate the decision.

Figure 3.2 steps in decision making process

3.41 Step 1: Problem identification and diagnosis

The first stage of the decision-making process is recognising that a problem exists and that action has to
be taken. A problem is a discrepancy between the current state of affairs and a desired state of affairs.
Unless the problem is identified in precise terms, solutions are very difficult to find. In seeking to identify
a problem, managers can use a variety of sources of data, including comparing organisational performance
against historical performance, against the current performance of other organisations/departments or
against future expected performance. Problem identification must be followed by a willingness to do
something to rectify the situation. Before taking action the problem needs accurate diagnosis. Diagnosis
involves assessing the true cause of the problem by carefully selecting all relevant material and discarding
information which is not relevant to the problem at hand.
3.42 Step 2: Identification of alternatives

Having identified and diagnosed the problem, the next step for an organisation is to identify a range of
alternatives to solve the problem. Managers should try to identify as many alternatives as possible in order
to broaden options for the organisation. In generating alternatives the organisation may look toward ready-
made solutions that have been tried before, or custom-made solutions that have to be designed specifically
for the problem at hand. In today’s business environment more and more organisations are applying
custom-made solutions to enhance competitive advantage.

3.43 Step 3: Evaluation of alternatives

Having identified the available alternatives, a manager needs to evaluate each alternative in order to
choose the best one. Consideration should be given to the advantages and disadvantages as well as the
costs and benefits associated with each option. Most alternatives will have positive and negative aspects
and the manager will have to try to balance anticipated outcomes. Depending on the situation, evaluation
of alternatives may be intuitive (based on gut feeling) or based on scientific analysis. Most organisations
try to use a combination of both. When evaluating alternatives, managers may consider the potential
consequences of alternatives under several different scenarios. In doing so they can develop contingency
plans which can be implemented with possible future scenarios in mind.

3.44 Step 4: Choice of alternative

Having evaluated the various alternatives, the next step is to choose the most suitable one. If for some
reason none of the options considered is suitable, the manager should revert back to Step 2 of the process
and begin again. When there are suitable alternatives and Steps 2 and 3 have been conducted skilfully,
selecting alternatives may be relatively easy. In practice, however, alternatives may not differ significantly
in terms of their outcomes and therefore decisions will be a matter of judgement. In coming to a decision
the manager will be confronted by many conflicting requirements that will have to be taken into account.
For example, some trade-offs may involve quality versus acceptability of the decision, and political and
resource constraints.

3.45 Step 5: Implementation

Once the decision has been made it needs to be implemented. This stage of the process is critical to the
success of the decision and is the key to effective decision making. The best alternative is worth nothing
if it is not implemented properly. In order to successfully implement a decision, managers must ensure
that those who are implementing it fully understand why the choice was made, why it is being
implemented, and are fully committed to its success. Decisions often fail at the implementation stage
because managers do not ensure that people understand the rationale behind the decision and that they are
fully committed to it. For this reason many organisations are attempting to push decision making further
down the organisation to ensure that employees feel some sense of ownership in the decisions that are
made.

3.46 Step 6: Evaluation

Once the decision is implemented, it needs to be evaluated to provide feedback. The process of evaluation
should take place at all managerial levels. This step allows managers to see the results of the decision and
to identify any adjustments that need to be executed. In almost all cases some form of adjustment will be
made to ensure a more favourable outcome. Evaluation and feedback are not one-off activities, however,
and they should form part of an ongoing process. As conditions change, decisions should be re-evaluated
to ensure that they are still the most appropriate for the organisation. This also helps managers to learn
about making sound decisions taking past experience into account.

3.5 Approaches to decision making

The four most popular approaches to the study of decision making are: the rational model; bounded
rationality; the political model; and escalation to commitment.

3.51 The concept of rationality

Rationality in relation to decision making refers to a process that is perfectly logical and objective,
whereby managers gather information objectively, evaluate available evidence, consider all alternatives
and eventually make choices that will lead to the best outcomes for the organisation. The rational approach
to decision making has its foundations in traditional economic theory, which argues that managers attempt
to maximise benefits and have the capacity to make complex decisions quickly. Such a rational approach
to decision making assumes that four conditions are fulfilled: 1. There is perfect knowledge of all the
available alternatives. 2. There is perfect knowledge of all of the consequences of the available
alternatives. 3. Managers have the capacity to objectively evaluate the consequences of the available
alternatives. 4. Managers have a well-structured and definite set of procedures to allow them to make
optimum decisions.

3.52 Bounded rationality

As we have seen, decisions are made under varying conditions ranging from certainty and risk to
uncertainty. In the current environment managers seldom make decisions under the conditions of certainty
that would be needed to apply a completely rational model. For many managers today the rational
approach represents an ideal approach, but one that is simply not attainable under current conditions of
risk and uncertainty. Given the fact that managers cannot always make decisions under certainty
conditions, and in a rational manner, they have to apply a less than perfect form of rationality. Herbert
Simon called this ‘bounded rationality’, and argued that decisions taken by managers are bounded by
limited mental capacity and emotions, and by environmental factors over which they have no control. Due
to these limitations’ managers rarely maximise or take ideal decisions with the best possible outcomes.
Intuition and judgement are therefore used by the manager to solve problems and make decisions. Taking
a rational approach to problem solving and decision making involves clear identification of goals,
objectives, alternatives, potential consequences and their outcomes. Each of these is in turn evaluated in
terms of contribution to the overall aim.

psychological biases can influence judgement. For example, a manager might have to make a decision
about where to establish a subsidiary office of the organisation. When making the decision the manager
could be influenced by personal opinions, emotions and personal bias in favour of one location over
another. This might be particularly noticeable if the manager is subsequently going to work in the office,
as the choice might be heavily influenced by his/her desire to live in one location. In this way, total
rationality is not applied as the manager may choose a location that s/he favours and this will not
necessarily be the most rational choice.
Managers may also be unable to handle large amounts of complex information. Bounded rationality also
recognises that managers may not have full and complete information and may experience problems
processing information, which clearly affects a manager’s ability to make optimal decisions.12 Decisions
made under bounded rationality may not always be the best; however, on occasion good decisions have
been made on the basis of judgement and gut feeling.

3.53 The political model

While the previous approaches have concentrated on the role played by rationality in the decisionmaking
process, the political model concentrates on the impact of organisational politics on decision making.
Power and politics play an important role in the decision-making process. Power is the ability to influence
others. In the context of an organisation power can be viewed as the ability to exert influence over
individuals, work groups or departments.

There are five main types of power found in the organisational setting:

1. Legitimate power originates from the manager’s position within the organisation’s hierarchy. The
power is inherent in the hierarchical position the manager occupies.
2. Reward power originates from the manager’s ability to withhold rewards from others.
3. Expert power derives from the expert knowledge and information that an individual/manager has
amassed.
4. Referent power originates from the charisma or identification that a manager has developed.
5. Coercive power is associated with emotional or physical threats to ensure compliance.

In the decision-making process those who possess power are clearly an important dynamic. Political
decision processes are used in situations where uncertainty, disagreement and lack of information are
common. Within organisations it is common to find different coalitions, all of which possess varying
degrees of power depending on the situation. Coalitions can be formed by particular work groups, teams,
managers, functional specialists, external stakeholders and trade unions. Each group brings with it certain
ideas and values, coupled with power, in relation to the decision under discussion. It is common for each
coalition to defend its own territory and to ensure that any decisions made do not negatively impact on its
members (both formal and informal). The presence of coalitions therefore adds an important ingredient to
the decision-making process.

3.54 Escalation of commitment

While it does not explain how decisions are made, this approach concentrates on why people continue to
pursue a failing course of action: that is, why commitment to a poor decision often escalates after the
initial decision has been made. This approach is particularly concerned with decision makers who, even
in the face of failure, continue to invest resources in a failing decision. For example, an organisation may
decide to enter a particular market by introducing a certain product. After a little while it may become
obvious that the product is not suited to that market. The organisation, however, continues to increase
spending on advertising and marketing rather than exiting from the market. Escalation of commitment to
a failing decision is often attributed to self-justification and a feeling of personal responsibility for the
decision. When individuals are personally responsible for negative consequences they may decide to
increase investment of resources in a previously chosen course of action.13 Organisations therefore have
to strike a balance between persevering with a decision and recognising when a decision is failing and
should be abandoned. Not all organisations fall into the escalation of commitment trap.

3.6 Group versus individual decision making

Task forces, teams and boards are all examples of where decision making occurs in a group setting. The
basic idea behind group decision making is the notion that two heads are better than one. Generally the
diversity of groups facilitates better-quality decisions.14 However, a group can be inferior to the best
individual in the group.15 In some cases, groups will provide the best-quality decisions and in others the
individual will do better.

Group decision making: advantages and disadvantages Advantages

i. Group decision making allows a greater number of perspectives and approaches to be considered,
thereby increasing the number of alternatives that can be drawn up.
ii. Groups generally facilitate a larger pool of information to be processed. Individuals from different
areas can bring varied information to the decision-making setting.
iii. By increasing the number of people involved in the process it is more likely that a greater number
of people will understand why the decision was made, and this facilitates implementation.
iv. Group decision making allows people to become involved and produces a sense of ownership of
the final decision, which means that people will be more committed to the decision.
v. Using a group to arrive at a decision means that less co-ordination and communication is required
when implementing the decision.

Disadvantages

i. Group decisions take longer to arrive at and this can be problematic when speed of action is key.
ii. Groups can be indecisive and opt for satisficing rather then maximising. Indecision can arise from
lack of agreement among members. Satisficing occurs when individuals grow tired of the process
and want it brought to a conclusion, leading to satisficing rather than maximisation.
iii. Individuals who have either a strong personality or a strong position can dominate groups. The
result is that a particular individual can exert more influence than others. The main problem with
such a situation is that the dominating person’s view of the decision need not necessarily be right;
and if his/her view is the right one, convening a group for discussion is a waste of time.
iv. Groups inevitably have to compromise to reach a decision and this can lead to mediocre decisions.
Mediocrity results when an individual’s thinking is brought into line with the average quality of a
group’s thinking. This is called the levelling effect.
v. Groups can lead to group think, which can be defined as ‘a mode of thinking that people engage
in when they are deeply involved in a cohesive group, when members’ strivings for unanimity
override their motivation to realistically appraise alternative courses of action’. Group think
happens in situations where the need to achieve consensus among group members becomes so
powerful that it takes over realistic evaluations of available alternatives. Criticism is suppressed
and conflicting views are not aired for fear of breaking up a positive team spirit. Such groups
become over confident and too willing to take risks.

3.7 Improving group decision making

In order to avoid the disadvantages associated with group decision making and to build on the advantages,
three main ways of improving group decision making have been proposed.

a) Brainstorming

Brainstorming became popular in the 1950s, was developed by Alexander Osborn to facilitate the
development of creative solutions and alternatives. Brainstorming is solely concerned with idea
generation rather than evaluation, choice or implementation. The term effectively means using the
brain creatively to ‘storm’ a problem. It is based on the belief that when people interact in a relaxed
and unrestrained setting they will generate creative ideas. The acceptance of new ideas is also more
likely when the decision is made by the group involved with its implementation.18 In brainstorming
the group members are normally given a summary of the problem before the meeting. At the meeting
members come up with various ideas, which are recorded in full view of all other members. None of
the alternatives is evaluated or criticised at this stage. As members produce new ideas and alternatives
this serves to stimulate other members in the hope that a truly good solution can be identified.

b) The Delphi technique

The was developed in the early 1960s as a means of avoiding the undesirable effects, while retaining
the positive aspects, of group interaction.19 Delphi was the seat of the Greek god Apollo, who was
renowned for his wise decisions. The Delphi technique consists of a panel of experts formed to
examine a problem. Rather than physically meeting, the various members are kept apart so that social
or psychological pressures associated with group behaviour cannot influence them. In order to find
out their views, they are asked to complete a questionnaire. A coordinator then summarizes the
findings and members are asked to fill out another questionnaire to re-evaluate earlier points. The
technique assumes that, as repeated questionnaires are conducted, the range of responses will narrow
to produce a consensus. The Delphi technique is particularly useful where experts are physically
dispersed, anonymity is required and members have difficulty communicating with each other. On the
negative side, however, it reduces direct interaction among group members.

c) Nominal grouping

This was developed in the 1970s. In contrast to brainstorming, it does not allow a free association of
ideas, tries to restrict verbal interaction and can be used at many other stages of the decision-making
process apart from idea generation.20 In nominal grouping, members are given a problem and are
asked to think of ideas individually with no discussion. They then present these ideas on a flip chart.
A period of discussion follows, which builds on the ideas presented. After the discussion, members
privately rank the ideas. Generation of ideas and discussion proceeds in this manner until a solution is
found.21 The main advantage of this approach is that it overcomes differences in power and prestige
between members and it can also be used at a variety of stages in the overall decision-making process.
Its main disadvantages are that its structure may limit creativity and it is costly and timeconsuming.