Behavioural economics: bounded rationality, heuristics and biases

Detail

1. From homo economicus to real people

• Homo economicus means "economic man". It is the textbook model of a person who is fully rational and fully selfish. This person knows every option, calculates perfectly and always picks the best one.
• Behavioural economics combines psychology and economics. It studies how biases, heuristics and social norms push real decisions away from strict rationality.
• Bias: an error in judgement that repeats in a set pattern. It is not a random mistake.

• Social norm: what people around us do, or expect us to do.

• Thaler's work showed that human limits and biases shape economic decisions in a large and systematic way, for both individuals and institutions. This goes against the standard assumption that people always act rationally and selfishly [7].

• Why "systematic" matters for the exam: random errors cancel out across a large population, but systematic errors do not. They add up and move markets and national outcomes, such as savings rates, asset bubbles and tax compliance.

2. Bounded rationality and satisficing (Herbert Simon, Nobel 1978)

• Bounded rationality: people are rational only within limits of:
• information: they do not know every option;
• time: decisions have deadlines;

• brainpower: the mind cannot compare hundreds of options.

• So people satisfice. Satisficing (satisfy + suffice) means picking the first "good enough" option instead of searching for the best one.

• Example: a job-seeker accepts the first offer above ₹30,000 a month and does not compare all 50 openings.

• Contrast with the neoclassical model: the neoclassical consumer maximises utility (satisfaction), while Simon's consumer satisfices.

3. Heuristics and biases (Kahneman–Tversky, 1974)

• Heuristics are mental shortcuts, or rules of thumb. They save effort, but they cause systematic biases.

(a) Anchoring

• Anchoring means leaning too heavily on the first number you see, called the "anchor".
• Example 1: a high printed MRP (maximum retail price) makes any discount look like a bargain.

• Example 2: a "was ₹999, now ₹499" tag. The buyer judges the deal against ₹999, not against the product's real worth.

• Chain: a high anchor is shown → the buyer's sense of "fair price" moves up → they are willing to pay more.

(b) Availability heuristic

• The availability heuristic means judging how likely something is by how easily examples come to mind.

• Example: people fear flying after a widely reported air crash, even though road travel is riskier.

• Chain: heavy media coverage → the event is easy to recall → people think it is more likely than it is → their risk choices go wrong.

(c) Confirmation bias

• Confirmation bias means looking for information, and reading it, in ways that confirm what one already believes.
• Example: an investor who likes a stock reads only positive reports about it and ignores warnings.

(d) Overconfidence bias

• Overconfidence bias means overrating one's own knowledge or forecasting skill.

• Example: retail traders in derivatives. Derivatives are contracts, such as futures and options (F&O), whose value comes from an underlying asset like a share or an index.

• Indian evidence (SEBI studies):

• About 9 out of 10 individual traders in equity F&O made net losses in both FY 2018-19 and FY 2021-22 [4].
• 93% of individual traders made losses in equity F&O between FY22 and FY24. Their total losses were more than ₹1.8 lakh crore over those three years [2][3].
• 7 out of 10 individual intraday traders in the equity cash segment made losses (SEBI study, July 2024) [5].
• Most traders keep trading despite repeated losses. This is overconfidence at work, often together with herd behaviour and the sunk cost fallacy.

4. Prospect theory (Kahneman–Tversky, 1979; Kahneman Nobel 2002)

• Prospect theory says people judge outcomes as gains or losses measured from a reference point, such as their current salary or the price they paid. They do not judge outcomes by their final total wealth.
• Standard (expected utility) theory differs: only final wealth matters there, so a ₹1,000 gain and a ₹1,000 loss are simply opposite moves.
• The 2002 Nobel is seen as the prize for behavioural economics, meaning the use of psychology in economic theory [8].

(a) Loss aversion

• Loss aversion means a loss hurts more than an equal gain pleases. Losses feel about 2 times as large as gains.
• Simple rule: pain of a loss of ₹X ≈ 2 × pleasure of a gain of ₹X. The factor 2 is called the loss-aversion coefficient, λ (lambda).
• Worked example: a coin toss. Heads, you win ₹150. Tails, you lose ₹100.
• Expected value = 0.5 × 150 − 0.5 × 100 = +₹25. A fully rational person accepts the bet.
• Felt value = 0.5 × 150 − 0.5 × (2 × 100) = 75 − 100 = −25. A loss-averse person refuses.

• They accept only when the win is more than ₹200.

• Policy use: framing a penalty as a loss works better than offering the same amount as a reward. For example, "you will lose your ₹500 deposit" works better than "you will earn ₹500".

(b) Endowment effect

• The endowment effect means people value something more just because they own it.

• Example: a farmer will not sell inherited land at the market price, yet would never buy the same land at that price.

• It comes from loss aversion. Selling feels like a loss.

(c) Framing effect

• The framing effect means the same information, presented in a different way, changes people's choices.

• Example: patients accept surgery far more readily when told "90% survival" than when told "10% mortality". The facts are identical.

• Policy use: the Economic Survey proposed moving from "Give it up" for the LPG subsidy to "Think about the Subsidy". The idea is to reframe the subsidy so that better-off households choose to leave it [6][9].

5. Thaler's anomalies (Richard Thaler, Nobel 2017)

• An anomaly is behaviour that standard theory cannot explain.
• Thaler's 2017 prize extended Kahneman–Tversky's work into policy-making. It is described as the prize for behavioural public policy [8].
• He co-wrote "Nudge: Improving Decisions about Health, Wealth, and Happiness" [7].

(a) Mental accounting

• Mental accounting means treating money differently depending on where it came from or what it is for, as if money were not fungible.
• Fungible means every rupee can be swapped for any other rupee.
• Example: a bonus is splurged while salary is saved.
• Worked example: a person keeps ₹50,000 in a savings account earning 3%. At the same time, they carry ₹50,000 of credit-card debt at 36%. Rationally, they should pay off the debt and save ₹16,500 a year (33% of ₹50,000). Their "mental accounts" stop them.

(b) Present bias

• Present bias means giving too much weight to rewards now compared with rewards later.
• Chain: "enjoy today" wins → saving is postponed → procrastination and under-saving for retirement.

(c) Hyperbolic discounting

• Discounting means valuing future money less than money today.
• Exponential (rational) discounting uses one fixed rate:
• PV = A / (1 + r)ᵗ
• PV = present value; A = amount; r = discount rate; t = time.

• Preferences stay the same over time.

• Hyperbolic discounting: the discount rate falls as the delay grows, so preferences are time-inconsistent (they change as time passes).

• PV = A / (1 + k·t)

• k = impatience parameter. The value falls steeply for short delays and gently for long delays.

• Worked example (from the scaffold):

• People take ₹100 today over ₹110 tomorrow. So waiting 1 day "costs" more than 10%.
• Yet they take ₹110 in 31 days over ₹100 in 30 days. The same 1-day wait now seems worth less than 10%.

• When day 30 actually arrives, the same person switches back to taking ₹100.

• Result: under-saving for retirement and delayed loan repayment.

(d) Status quo bias

• Status quo bias means preferring the current state of things even when better options exist.
• This is why defaults are so powerful (Section 8). A default is the option that applies if you do nothing.
• The Economic Survey 2018-19 noted that people suffer from great inertia (reluctance to act or change) and tend to stick to the default. So changing the default on an enrolment form, which costs almost nothing, uses this inertia for people's own good [9].

(e) Sunk cost fallacy

• A sunk cost is past spending that cannot be recovered.
• The sunk cost fallacy means continuing something because of past spending, instead of judging only future costs and benefits.
• Example: pouring more money into stalled infrastructure projects.
• Worked example: a project has used ₹800 crore. Finishing it needs ₹500 crore more and will return ₹300 crore. The rational choice is to stop, because 300 < 500 and the ₹800 crore is gone either way. The "we have already spent ₹800 crore" argument is the fallacy.

(f) Herd behaviour

• Herd behaviour means following the crowd and ignoring one's own information.
• Chain: others are buying → I buy too → prices rise further → more people join → a bubble forms → it bursts.
• It fuels asset bubbles, bank runs (depositors rushing to withdraw because others are withdrawing) and IPO frenzies (huge oversubscription of new share issues).

6. Behavioural finance vs the efficient-market hypothesis

• The efficient-market hypothesis (EMH) says share prices already reflect all available information. So no one can beat the market regularly.
• Behavioural finance (Shiller, Thaler) uses these biases to challenge EMH. It argues that:
• overconfidence and herding push prices away from their true value (bubbles and crashes);

• loss aversion makes investors hold losing shares too long and sell winners too early.

• The SEBI loss data on F&O traders is Indian evidence that individual investors are not the fully rational actors the model assumes [2][3].

7. From theory to Indian policy: the "nudge"

• Nudge policies are a new type of policy that sits between laissez-faire and incentives. Laissez-faire means leaving people fully alone; incentives means paying or fining them.
• Nudges use human psychology to change the choice architecture, which is the way options are presented to people.

• They gently steer people towards good behaviour and still leave them free to choose [9].

• Economic Survey 2018-19, Chapter "Leveraging the Behavioural Economics of 'Nudge'":

• The Survey's theme was "Shifting Gears" towards #Economy@5trillion, including behavioural "nudges" to design and deliver better outcomes [10].
• It cites the 2017 Nobel to Richard Thaler as the peak of the field's progress [10].
• Success stories named: the Swachh Bharat Mission (SBM) and Beti Bachao Beti Padhao (BBBP) [6].

• #SelfieWithDaughter became a worldwide hit. Celebrating the girl child became a social norm that more people wanted to follow [6].

• Agenda proposed by the Survey [6]:

• from BBBP to BADLAV (Beti Aapki Dhan Lakshmi Aur Vijay Lakshmi);
• from Swachh Bharat to Sundar Bharat;
• from "Give it up" for the LPG subsidy to "Think about the Subsidy";

• from tax evasion to tax compliance.

• Limits: nudges cannot and should not replace every incentive-based or mandate-based policy. Serious harms need strict rules or a stronger push than a nudge [9].

Prelims Hooks

• Bounded rationality and satisficing: Herbert Simon, Nobel 1978. Satisficing means choosing "good enough", not the best.
• Heuristics and biases paper: Kahneman–Tversky, 1974. Prospect theory: 1979. Kahneman's Nobel: 2002. Thaler's Nobel: 2017 [7].
• Loss aversion: losses weigh about 2× equal gains. Endowment effect: people overvalue what they own. Both come from prospect theory, not from Thaler's list.
• Framing trap: "90% survival" vs "10% mortality" carries the same facts but leads to different choices.
• Hyperbolic discounting: the discount rate falls with delay, so preferences are time-inconsistent. Exponential discounting uses a constant rate.
• Mental accounting breaks the idea that money is fungible. Sunk cost fallacy means letting unrecoverable past costs drive decisions.
• Status quo bias is why defaults (opt-out designs) work [9].
• The Economic Survey chapter on "Nudge" appeared in 2018-19. It named SBM and BBBP as successes and proposed BADLAV [6][10].
• SEBI: 93% of individual equity F&O traders made losses between FY22 and FY24, with total losses above ₹1.8 lakh crore [2].
• Trap: behavioural finance challenges the efficient-market hypothesis. It does not support it.

Mains Points

• Nudge as low-cost governance (GS-III/GS-II):
• Changing defaults and using social norms costs little and keeps people free to choose.
• Examples: SBM, BBBP (#SelfieWithDaughter), "Give it up" [6][9].

• Limits: nudges cannot replace mandates where the harm is serious [9]. Critics also call nudges paternalism, meaning the state deciding what is good for people.

• Retail investor protection:

• Overconfidence, herding and loss aversion explain why 93% of individual F&O traders lost money between FY22 and FY24 (above ₹1.8 lakh crore in total) [2].

• This supports SEBI-type action on disclosure and entry barriers, alongside financial literacy.

• Savings and pensions:

• Present bias and hyperbolic discounting explain under-saving for retirement.

• Auto-enrolment and default contributions in pension schemes can raise the household savings rate, which funds investment.

• Public investment discipline:

• The sunk cost fallacy keeps stalled projects alive.
• Ongoing evaluations based only on future costs and benefits, and time-bound exit rules, improve capital efficiency.