Behavioural Policy Design Week 6

Question 1

What is most policy about

2. 2 traditional policy interventions

Answer 1

Inducing behaviour change e.g stop speeding

2. Price based or quantity based, now we have behavioural based also

Question 2

2 parables on behaviour

Answer 2

The mystery of missing parents

The legend of Jos Van Bedaf

Question 3

The mystery of the missing parents- what happened?

Answer 3

Parents picking up children late.

Fined late-comers.

Actually increased amount of late-comers.

Question 4

2 policy interventions to control behaviour

Answer 4

Quantity-based
Price-based

Question 5

Why did late-comers increase?

Answer 5

Feel less guilty, as time being paid for through fines.

Becomes a transaction.

Question 6

Main takeaways from “the mystery of the missing parents”

Answer 6

Be more cautious with model-based predictions.

Is it extrinsic vs intrinsic motivation? One crowds out the other. In this case=extrinsic motivation being the fine.

Question 7

Extrinsic vs intrinsic motivation

Answer 7

Extrinsic e.g fines, penalties etc

Intrinsic e.g being a good person, moral code

Question 8

Parable 2: Legend of Jos van Bedaf

Answer 8

Problem of dirty toilets; creates budgetary and public health issues.

Could have a inspector but awkward and expensive.

Target in urinal (fly-sticker-intuition)

Question 9

How much did spillage reduce by? And how much money would this save a year?

Answer 9

85%.

250k euro

Question 10

What is this fly-sticker an example of?

Answer 10

Choice architecture

Question 11

Choice architecture and what does it require?

Answer 11

Controlling behaviour by design of environment

Requires an atheoretical approach- trial experiments

Question 12

Atheoretical approach

Answer 12

Encourage use of trial experiments. Less theory-driven

Question 13

Private sector nudges in supermarkets examples (2)

Answer 13

Key goods at eye level

seeing fresh food first and snacks last when waiting in queue.

Question 14

Benartzi et al- More applications of behavioural policy (4)

Answer 14

Norms
Defaults e.g people with drivers license immediately opted in for organ donation
Reminders
Triggering Human instinct e.g brains attracted to symmetry so exploit this!

Question 15

Example experiment exploiting human instinct to symmetry

Answer 15

Thermostat experiment shows people return thermostat back to symmetrical, and so designed the temperature metrics to align symmetrical at the temperature that is desired by the company

Question 16

Pros of behavioural policy design (2)

Answer 16

Promise of big behaviour changes in exchange for small/cheap changes to choice architecture. Think the 2nd parable.

Nudges often cheaper than traditional interventions (Benartzi) Ratio of impact to cost better!

Question 17

Nudges often cheaper than traditional interventions example figure.

KEY: Nudges are COST-EFFECTIVE

Answer 17

For college enrolment, increased by 1.53 students per $1000 spent on nudges. (Bettinger et al, 2012)

Compared to 0.0051 students per $1000 spent on monetary subsidies (Long, 2004)

Question 18

Cons/challenges of behavioural policy design (5)

Answer 18

-Behavioural heterogeneity makes policy designing different i.e People respond differently

-Behavioural policy interacting with pre-existing policy can be complicated i.e multiple instruments=hard to understand

-Longevity-robustness-‘tricks’ people stop following ones especially that involve trickery. Maybe only can be done one-off, people go back to what they were doing before

-Welfare-paternalism-normative analysis becomes hard to do. Removes revealed preference, choice architect makes the decisions. Hard to know ‘good and bad’ outcomes. (Dan Ariely’s shows range of choice is more limited than we think)

-The most cost effective nudges do not always create LARGE changes in behaviour, so benefit undermined. (Case follows!)

Question 19

Who established the nudge unit first, with others following?

Answer 19

UK in 2010, incorporating behavioural science into policy.

Question 20

Nudge definition (if needed) from Thaler.

Answer 20

An intervention easy and cheap to avoid. Not mandates, do not change economic incentives or forbid options.

Question 21

Default examples in Benartzi (3) and defaults are effective? And who says this

Answer 21

Automatically opt in people with drivers licenses for organ donation
For programmes to reduce poverty
To reduce paperwork requirements for obtaining licenses or permit

Efffective as people infer policymakers are recommending the default option, thus becoming the reference points, and deviating is different. (Johnson and Goldstein)

Question 22

Linking back to one of the disadvantages of behavioural policy: the most cost-effective nudges are undermined as don’t have large shifts in behaviour

Key case: DOD and SBST

Answer 22

Emails to increase savings among military.

Emails resulted in higher enrolment rate 1.6-2.1%, compared to 1.1% rate with no emails.

Appears a small shift in behaviour: but calculated to increase savings by $8M approx in total in a year.

Question 23

Milkman et al.

What % did it increase by?

Answer 23

Looked at health- nudge to increase influenza vaccinations.

Prompts would remind people to plan the date and time of vaccinations, to embed in memory and create commitment uncomfortable to break.

Increased vaccinations by 4.2%

Question 24

Chapman studied vaccination rates too…
Automatically opted in, increased by how much

Answer 24

11.7%

Question 25

So when are nudges better (2) , and when are monetary incentives (that shift economic incentives) better

Answer 25

Monetary incentives better when correcting misalignment between public interest and private interests of citizens.

Nudges better in changing day to day behaviour of individuals who are making imperfect decisions e.g biased, rushed etc.

Nudges also work in concert with other tools e.g the savings plan, the vaccinations, as opposed being effective in policy involving externalities e.g preventing violent crime.