Beyond Rationality

Dr. Divya Gupta

Economics I

Dr. Divya Gupta Beyond Rationality Economics I 1 / 22

References

Neva Goodwin: Chapters 8 and 9

Class Discussions

Dr. Divya Gupta Beyond Rationality Economics I 2 / 22

Philosophical Premise: Methodological Individualism

Neo-classical economics focuses on individual behaviour - individual firm or

individual consumer.
In micro economics, we are trying to understand these individual behaviors
and actions, that is, demand and supply.
Hence to model these actions, we look at consumer theory and firm theory.
How does such a micro-level analysis fit in the larger scheme of things -
answering the broad and important question of ‘allocation of limited
resources’ ?
The philosophical premise on which the neoclassical models are based is
called “Methodological individualism” - the idea that ‘social phenomena are
results or outcomes of individual actions or decisions’.
Thus, we study these individual actions as a step towards understanding the
broad social phenomenon.
If we can ensure that these individual decisions are taken rationally and in an
efficient manner, we can ensure efficiency of the social outcome.

Dr. Divya Gupta Beyond Rationality Economics I 3 / 22

A sidenote: Critique of Methodological Individualism

Have to think of a meaningful way of aggregating individual

preferences into a social phenomenon.
Should individual choices be simply added or with weights? What
weights should be attached? etc.
What considerations need to be taken into account while aggregating?
By focusing on the social outcome per se, and not on what ‘individual
decisions’ make those outcomes, this philosophy ignores interpersonal
comparisons.
This philosophy says that social norms are results of individual actions
and not vice-versa.
However, in reality, the reverse is also true. That is, individual actions
are also affected by social norms (recall the example of herd
behaviour to pursue an MBA degree after [Link]).

Dr. Divya Gupta Beyond Rationality Economics I 4 / 22

Back to Methodological Individualism

Let’s ignore the critiques of methodological individualism and let’s take the
discussion forward with this as a foundational philosophy.
Although as economists/ social scientists, we are really interested in solving
the problem of the society, we start with individual utility. Why?
Note that our main motive as an Economist is to understand the market
mechanism. The idea is to understand the free market allocation; simply
put, to understand the equilibrium prices and quantity, that are determined
as a result of perfect competition.
Therefore, these Individual decisions and choices are mere causal
intermediaries.
Economists are more interested in the market consequences of these
individual choices than in the individual choices themselves.
As per the Rational Choice Theory, the starting point for modelling these
individual choices is ‘rational preferences’.

Dr. Divya Gupta Beyond Rationality Economics I 5 / 22

Methodological Individualism (contd.)

Therefore, we have something like this:

Individual preferences → Individual Choices (individual demand) →
(aggregated - based on methodological individualism) → social phenomenon
(free market outcomes of equilibrium P and Q).
While modelling individual decisions, we start with certain assumptions
about these preferences so as to give us rational preferences.
We, now, need to have a closer look at these four axioms of preferences:
1 Completeness - always required for existence of an ordering of preferences.
2 Transitivity - always required for internal consistency of preferences.
3 Reflexivity - trivial (almost always satisfied) .
4 More is Better - holds true until the consumer reaches the point of satiation.
Not true beyond the point of satiation.
Therefore, we focus only on the first two axioms which are the most
important axioms to be satisfied to establish rationality of preferences.
Next we critically discuss, step-by-step, the rational choice theory or
simply put, the utility theory.
Dr. Divya Gupta Beyond Rationality Economics I 6 / 22
I. Axioms of Completeness and Transitivity

As discussed previously, out of the four axioms of preferences, the

axioms of ‘reflexivity’ is trivial, and that of ‘more is better’ will hold
true under situations of non-satiation.
Now, if we focus on just the other two axioms of ‘completeness’ and
‘transitivity’, these are very hard to be satisfied in reality.
It is not always possible for an individual to ‘know their preferences’
or be always decisive. In fact, in situations of uncertainty (as we will
discuss later), they do not even know of all possible alternatives.
Hence, axiom of completeness is hard to be satisfied in reality.
Example: Remarks like ‘I don’t know’, ‘I’m confused’, or ‘not very sure’
indicate an individual’s indecisiveness.
Likewise, individuals do not always display consistent preferences.
They may prefer a particular alternative in a particular situation and
may completely dislike it in another situation, hence, axiom of
transitivity is also difficult to be satisfied in real situations.
Dr. Divya Gupta Beyond Rationality Economics I 7 / 22
II. Assumption of Self-interest Maximisation
Now, suppose we ignore the first point of critique.
Let’s assume that the two axioms of ‘completeness’ and ‘transitivity’ are actually
the ‘minimalistic’ requirements for rationality, because all these require is for the
preferences to ‘exist’ and be ‘internally consistent’.
Thus, even if they are not always satisfied, they are not hard to be satisfied.
The economic theories employ these models/Utility Theory to explain free market
mechanisms, which requires an additional assumption of ‘self-interest
maximisation’.
This makes it a substantive model.
Further, it is not necessary for rationality to always imply ‘self-interest’.
However, the Utility Theory is much broader in scope than the limitations imposed
by the economic theories.
In fact, the model simply says that an individual will always choose the ‘best’
alternative from all possible set of outcomes.
This ‘best’ outcome need not necessarily imply self-interest maximisation, but only
maximisation of utility.
Thus, a philanthropist, for example, will be maximising their utility by maximising
others’ interests, rather than their own.
Thus, the Utility Theory can very much incorporate decisions taken out of love,
jealousy, altruism, etc.
Dr. Divya Gupta Beyond Rationality Economics I 8 / 22
III. Situation of Uncertainty
Lastly, and most importantly, even if we proceed with ignoring the first two
points of critiques, we get stuck into a deeper problem.
What happens to these axioms when there is incomplete information and
uncertainty - which is most often the case in the reality.
Uncertainty - when individuals are not certain about the consequences of
their actions.
Example: deciding which career path to choose. There could be different
career options like joining a firm as a Legal Advisor or practicing as a lawyer,
or prepare to join the civil services. Would you surely know the consequences
of each of these options?
Additionally, most of the times, even all possible outcomes are not known
and/or the probabilities of their occurrences.
In such situations, an individual’s ‘rational choice’ will be determined by not
only ‘rational preferences’ but also ‘rational beliefs’.
Hence, while in situation of certainty, we had:
Rational preferences → Rational Choices (under Utility Theory) →
(aggregated - based on methodological individualism) → Social outcome
Dr. Divya Gupta Beyond Rationality Economics I 9 / 22
III. Situation of Uncertainty (contd.)
However now, under uncertainty, we have:
Rational preferences + Rational beliefs → Rational Choices (under Expected
Utility Theory) → (aggregated - based on methodological individualism) → Social
outcome
Beliefs - subjective probabilities that an individual forms about the occurrence of
different alternative events, due to the uncertainty of those events.
Expected Utility Theory - where individuals ought to maximise their ‘expected
utility’ from different possible alternatives and their respective probabilities of
occurrences.
Rationality of beliefs is even more demanding and controversial to be satisfied in
the real life.
To establish rationality of beliefs, not only do we need the axioms of completeness
and transitivity to be satisfied, but in addition, these beliefs also need to satisfy
the axiom of ‘independence’ and all mathematical laws of probabilities.
Independence and mathematical laws of probabilities are the hardest to be satisfied
for most people and are more often violated, because individuals are affected by
‘cognitive biases’ arising out of ‘heuristics’.

Dr. Divya Gupta Beyond Rationality Economics I 10 / 22

Heuristics and Coginitive Biases
Heuristics - processes used to reduce a complex task of evaluating problems
to judgemental operations.
These are mental tools/ cognitive rule of thumbs/ mental shortcuts/
pattern recognition etc. used by individuals in their day-to-day
decision-making process to simplify them.
Example: when you place an order in a restaurant, based on what the
waiter said are the specials or simply what the other person ordered -
instead of going through the entire menu and objectively evaluating all
options.
Cognitive Biases - the biases that develop as a result of these heuristic
processes. That is, inclination towards a particular belief or prospective,
mostly one that is ill-supported by reason or evidence.
These biases, therefore, affect the formation of ‘subjective beliefs’ during
situations of uncertainty, which defy the mathematical laws of probability
and hence, violate the assumptions required for rationality of beliefs.
Next, we discuss such three types of cognitive biases as theorised and
empirically tested by Tversky and Kahneman (1974).
Dr. Divya Gupta Beyond Rationality Economics I 11 / 22
Types of Cognitive Biases

These cognitive biases have been discussed by Tversky and Kahneman in their
1974 paper - “Judgment under uncertainty: Heuristics and biases. science,
185(4157), 1124-1131”.
1 Representative Bias:
Refers to determining the probability of occurrence of an event by
incorporating how much is the event representative of the sample it is
drawn from.
That is, how much representative an event A is to the sample or how
‘similar’ A looks to the entire sample. If they are perceived to be very
similar, then the probability attached to occurrence of A is high,
otherwise, low.
These biases are often guided by “stereotypes” and people tend to
ignore the size of the sample while determining belief under such
considerations.
Recall the example of lawyers and housewives as discussed in the class.

Dr. Divya Gupta Beyond Rationality Economics I 12 / 22

Representative Bias/ Representative Heuristic

Dr. Divya Gupta Beyond Rationality Economics I 13 / 22

Types of Cognitive Biases (contd.)

2 Availability Bias:
Situations in which people asses the probability of an event by the ease
with which instances or occurrences can be brought to mind.
This happens due to the familiarity or the recency of an event.
Example: one may assess the risk of heart attack among middle-aged
people to be higher, by recalling such occurrences among one’s
acquaintances - which may not be true statistically.
Such biases often result due to ease of retrievability of instances, that
is, if information on occurrence of an event is easily available, it is
presumed to be a more probable occurrence.
Another factor that gives rise to such biases is the ease of imagination.
In situations of uncertainty, one often has to imagine the possible
outcomes and also the possibility of their occurrences. Therefore, easy
to imagine outcomes are assigned higher probabilities.
Recall examples discussed in the class.

Dr. Divya Gupta Beyond Rationality Economics I 14 / 22

Availability Bias/ Availability Heuristic

Dr. Divya Gupta Beyond Rationality Economics I 15 / 22

Types of Cognitive Biases (contd.)
3 Anchoring and Adjustment:
Series of mental shortcuts where quick decisions are taken by choosing a
reference point, and then adjusting to the reference point till we reach a
suitable decision.
Example - as the seller knows that there would definitely be bargaining by
the consumer, she would keep the price higher of the product say Rs 100.
Now, the consumer would automatically anchor to Rs 100, which becomes
their reference point. The consumer would, thus, adjust the bargaining to,
say Rs 70, and ultimately the equilibrium maybe reached at Rs 80.
On the other hand, if the seller would have priced the product at Rs 80
initially, then the equilibrium price could have been arrived at Rs 60.
Therefore, when we make decisions, we first determine an anchor and adjust
accordingly.

Dr. Divya Gupta Beyond Rationality Economics I 16 / 22

Types of Cognitive Biases (contd.)
3 Anchoring and Adjustment (contd.):
Different reference points/initial values may result in different estimates and
hence, different decisions.
Status Quo Bias/ endowment effect - Status quo bias describes
individuals’ preference for the current state of affairs; resulting in
resistance to change. This often results when the reference point/initial
value is high enough, that is, a high anchor point.
Aspirational Bias - Analogously, the aspirational bias describes
individuals’ preference towards what they aim to achieve, that is, for a
better state of affairs. This often results when the reference
point/initial value is very low, that is, a low anchor, and hence, giving
rise to a higher aspiration among individuals.
Status quo v/s aspiration: you score 23 out of 25 in a mid-semester
exam. 23 becomes your reference point and you will not want to
gamble it by giving a remedial exam (status quo).
However, if you score 10 out of 25, low reference point, you will want
to give a remedial (aspiration).
Dr. Divya Gupta Beyond Rationality Economics I 17 / 22
Status Quo Bias

Dr. Divya Gupta Beyond Rationality Economics I 18 / 22

Bounded Rationality
As a result of the various heuristics and biases, the decisions taken by
individuals become irrational; in such situations, an individual is said to be
‘bounded rational’.
Bounded Rationality was first proposed by Simon A. Herbert in 1955.
It means the decision-making that attempts to make sense of the world by
the way a person takes in information and processes it to ‘create’ preferences
and choices.
As per the theory of bounded rationality, individuals set for themselves a
level of aspiration based on:
limited/incomplete/inadequate information;
with limited cognitive ability, affected by heuristics and biases, to assess
the gains and losses; to reason; often accompanied by emotional
motivations; and
time constraints, to comprehend and dwell on all possible alternatives
and their outcomes (recall the restaurant example, where you are
rushed by the waiter to place an order!).

Dr. Divya Gupta Beyond Rationality Economics I 19 / 22

Bounded Rationality (contd.)

Based on the aspiration level set by individuals, given various

constraints, they choose the first alternative that meets their
aspirations, rather than choosing the ‘best outcome’ out of all
possible alternatives.
That is, they seek a ‘satisfactory solution’ rather than an ‘optimal
solution’.
Hence, individuals operating in the framework of bounded rationality
are said to be ‘satisficers’ (term coined by Simon), which is a
combination of two words - satisfy and suffice.
Individuals, therefore, are not ‘optimisers’ as is posited by the Utility
Theory/Rational Choice Theory, but they are satisficers.
Hence, in reality, rationality is believed to be ‘bounded’.

Dr. Divya Gupta Beyond Rationality Economics I 20 / 22

Rationality and Morality

Despite all the discussion above, even if we, for the time being,
believe that individuals indeed are ‘rational beings’, is rationality an
end goal in itself?
Does rationality necessarily imply morality? Think of a rational villain
- whose rationality would make him want to cause maximum amount
of pain! Think!

Dr. Divya Gupta Beyond Rationality Economics I 21 / 22

 ALL THE BEST!

Dr. Divya Gupta Beyond Rationality Economics I 22 / 22