ECON30019 Chap.16 Social Preferences and Behavioural Public Policy
Social Preferences and Behavioural Public Policy
Every model to this point took a single decision maker with preferences over their own outcomes. The self-interest assumption was never separately argued for, and it produces sharp corner predictions in any setting where one person's payoff can be raised by lowering another's: keep everything in an allocation design, offer the minimum in a bargaining design, contribute nothing in a contribution design.
All three predictions fail, and the failures are qualitative rather than marginal. Rejecting a positive offer is costly to the person rejecting it, which rules out confusion or indifference as explanations.
The chapter builds the two standard repairs, simple altruism and inequality aversion, shows why only the second explains rejection, and then turns to conditional cooperation, costly punishment and what a defensible behavioural policy intervention has to demonstrate.
What this chapter covers
- 01
Why this chapter is here, and what it is built from
- 02
The self-interest benchmark and its corner predictions
- 03
Three standard designs, and the shape of each failure
- 04
Why costly rejection rules out the easy explanations
- 05
Simple altruism: adding the other payoff with a weight
- 06
Why altruism cannot explain rejection
- 07
Inequality aversion: penalising the gap in both directions
- 08
The standard restriction on the two weights, and what it means
- 09
Conditional cooperation, unravelling, and costly punishment
- 10
Behavioural public policy: mechanism, cost and limit
Find the smallest acceptable offer
- 1Set up the payoffs: the responder receives the offer and the proposer receives the remainder of the pot.
- 1Note that for any offer below half the pot the responder is behind, so only the disadvantage term is active.
- 1Write the utility of accepting: the offer minus 0.6 times the gap between the two payoffs.
- 1Simplify: the offer minus 0.6 times twenty minus twice the offer, which is 2.2 times the offer minus 12.
- 1Write the utility of rejecting: both payoffs are zero and equal, so no inequality term applies and the value is zero.
- 1Solve and compare: accepting beats rejecting when the offer exceeds 12 over 2.2, approximately 5.45, whereas a self-interested responder accepts any strictly positive offer.
Key terms
- Self-interest benchmark
- The assumption that a decision maker's utility depends only on their own payoff, which produces corner predictions in every design where payoffs can be transferred.
- Simple altruism
- An objective that adds the other person's payoff with a positive weight. It explains giving but predicts that a positive offer is never rejected.
- Inequality aversion
- An objective that penalises the difference between payoffs, with a heavier weight on being behind than on being ahead, which is what allows costly rejection.
- Disadvantageous inequality
- The case of being behind the other party, penalised by the larger of the two weights in the standard restriction on the model.
- Conditional cooperation
- Contributing in response to what others contribute, which explains why contributions start high and unravel as low contributors pull the rest down.
- Costly punishment
- Paying a personal cost to reduce another party's payoff, which stabilises contributions at a high level rather than allowing them to unravel.
Social Preferences and Behavioural Public Policy FAQ
Why is inequality aversion preferred to simple altruism?
Because altruism cannot explain rejection. An altruist who is offered a small positive amount should accept it, since rejecting delivers nothing to either party and they value the other party's payoff as well as their own. Penalising the gap rather than rewarding the level allows a small offer to be turned down, because the disadvantage penalty can outweigh the small amount gained.
What is the standard restriction on the two weights, and why?
That the weight on being behind exceeds the weight on being ahead, and that the second stays below one. In words, being behind hurts more than being ahead, and nobody destroys their own payoff purely to avoid being ahead. Without the second condition the model would predict people burning money to level down, which is not observed.
What makes a behavioural policy intervention defensible?
Three things. It should be effective for a reason the models name rather than because one trial worked. It should preserve the choice set, so anyone who wants a different outcome can have it cheaply. And it should be transparent, in the sense that it still works when disclosed, which is what separates a default from a manipulation that depends on going unnoticed.
Exam move
Answer policy questions with a mechanism, a cost and a limit, in that order. The mechanism is the specific model from an earlier chapter that predicts the intervention will work; the cost includes the people whom the intervention pushes away from what they actually wanted; and the limit is the evidence about when the underlying effect is weak or absent.
That three-part shape uses the whole subject and is hard to write without having understood it. For the models, practise setting up the inequality-averse objective and solving for a threshold, since that is the one calculation this material supports.
Working through Social Preferences and Behavioural Public Policy in ECON30019? Sia is AskSia’s AI Economics tutor — ask any ECON30019 Social Preferences and Behavioural Public Policy question and get a clear, step-by-step explanation grounded in how ECON30019 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
