Lecture 2 (Obstacles to Change Part 1) Flashcards
What are the traditional economic assumptions of behavior?
Assumptions:
* Individuals know prices
* Individuals behave consistently over time
* Individuals predict their future preferences
* Individuals estimate correctly their own abilities
* Decisions are not impacted by emotions
* Decisions are not impacted by framing
* Decisions are not impacted by reference points
Example the human machine, reference to the Star Trek character Spock.
What about behavior in reality? Are the assumptions reasonable at any time? Is the misbehavior systematic repeating mistakes?
Assumptions vs. reality:
* Individuals don’t always know prices
* Individuals don’t behave consistently over time
* Individuals mispredict their future preferences
* Individuals overestimate their own abilities
* Decisions are impacted by emotions
* Decisions are impacted by framing
* Decisions are impacted by reference points
Reference to Homer Simpson
- In many settings assumptions are reasonable: people are mostly
self-interested and well-informed, with coherent goals and
reasonable skill at making plans. - But in other settings the standard assumptions are unreasonable
descriptions of behavior: they are systematically incorrect in
ways with important consequences.¨
People’s judgments and decisions are often far
from rational:
*they are affected by seemingly irrelevant factors
or fail to take into account important information.
Systematic mistakes:
*people fail consistently in the same type of
problem.
*People seem to be irrational in a predictable way
Example with optimal savings. People do systematic and predictable under-saving. That distance we se from optimal “saving” to observed “saving” is called bias.
How does noise and bias affect accuracy?
Imagine four shooting plates. They are called separate names, Accurate, Noisy, Biased and “Noisy and biased”.
In the accurate case, people shot really good a close to the middle of the plate. They have good information and act on it.
In the noisy case, people shot all over the plate. Their behavior is being affected of all kinds of things.
In the biased case, people should accurate but not in the middle. They have been affected by something, which lead them to believe that another point then the middle was the best.
In the noisy and biased case, people should to the biased site of the plate but all over the place. Here again they have been misnomered and mislead to believe something was right .
Does adding untraditional assumptions mean that we are abandoning traditional methods?
No
The implications of nontraditional assumptions, even those
involving limits to rationality, should be studied using standard
mathematical methods; tested using standard statistics and
econometrics; and judged by standard scientific criteria such as
parsimony, prediction, generality, insight.
Despite the limits to the empirical validity and applicability of
standard assumptions, they are often appropriate even if not
exactly right.
Behavioral economics should enhance, not replace, traditional
economics .
What are the differences of a doer and a planner from Stanovich and West (2000)?
Doer (behavioral economics)
System 1 (INTUITIVE)
* Fast
* Automatic
* Unconscious
* Effortless
* Associative
* Emotional
Planner (traditional economics)
System 2 (DELIBERATIVE)
* Slow
* Reflective
* Self-aware
* Effortful
* Deductive
* Logical
How does it affect you thinking fast or slow?
Faced with a lot of options or complex questions under timepressure, one would normally go with an easy or intuitive choice, which in some cases would be not the best option or a wrong answer.
Under time pressure you would normally go for the most intuitive answer, which activates system 1, but if we gave you more and more time, you would slowly gain the time you need to activate system 2, which is more deliberative.
An example of a game, where you have different amount of time to anser 10, 20 and 30 seconds. The more time you get, the higher the succes of getting correct answers. Therefore, we could argue that it is important to incorporate fast and slow behavior into our understanding.
How to change behavior, what are the core steps? Give a decription and example
OECD: “Basic” framework
- identify “B”ehaviour
- “A”nalyse biases
- design “S”trategies
- test “I”nterventions
- scale for “C”hange
An example applying BASIC to increasing enrollment in pension plans
- Behavior: Identify and better understand your policy problem: Increase pension savings by encouraging more citizens to enroll in pension plans.
- Analysis: Review the available evidence to identify the behavioral drivers of the problem: Individuals tend to stick with defaults and choose inaction over action
- Strategy: Translate the analysis to behaviorally informed strategies: Change the default. Automatically enroll individuals into pension plans and allow them to opt-out.
-Intervention: Design and implement an intervention to test which strategy best addresses the problem: Test whether allowing indivuals to opt.out increase pension savings rather then the current practice of opt-in.
- Change: Develop plans to scale and sustain behavior: Share results with citizens, apply findings to system-wide reminders and monitor longterm consequences of the intervention.
Impactually: BOOST model
- “B”ehaviour: What behavior do we want to change.
* Define the desired outcome
* Understand whose behavior
you want to change – the target
group
* Understand what behaviors will
lead to the desired outcome
* Be as specific as possible
- “O”bstacles: Identify barriers (bottlenecks) that keep the behavior from changing.
* What is happening today?
* What obstacles are restricting
the users from engaging in the
desired behavior? Five main categories of biases, Limited attention, Other regarding preferences, Overconfidence, Reference dependent preferences, Bounded self-control.
* If you don’t know yet, how can
you find out?
- “O”utline: Identify potential interventions to promote the behavior based on science.
* What nudge could tackle the
barriers?
* Take inspiration from
behavioral science? REFINE - different nudges are better to address specific biases.
* Remember: Make it easy!
- Study: Understand which intervention works best in our context by testing and quantifying the effect.
* Context is king – we need to find out what
works best in this particular situation
* An experiment helps isolate and quantify
the effect of the new feature
* A cost effective way to avoid rolling out a feature that doesn’t work.
- Tailor: Adapt the solution to your needs:
* If the results are positive: Great! Consider scaling
* If the results are negative: Don’t worry! We still learned a lot. What went wrong?
Elaborate on the case study with public transport in Skåne, Gravert amd Collentine 2018.
Behavior:
First it´s important to distinguish between outcome and behavior.
outcome - Reducce the environmental impact of transport in Skåne.
Behavior -
* Take the bike to work
* Carpool
* Public transport (we focus here)
* Drive less
* Switch to an electric car
Obstacle:
Second we need to understand the problem first.
What would a typical person do in this situation when considering taken public transport?
- Biased reasoning: Is public transport so much better than taking my car?
- Bounded self-control: Intentions vs. actual behavior. I want to do it, but is also just so easy to take my car. But is it actually and maybe it’s more expensive.
- Other regarding preferences, individuals care about what others think of them. They adhere to social norms and are concerned about fairness. What do other people do?
- Alternatives
-
Outline:
Third, what interventions could help in this case?
Nudge or economic incentive?
All the people in the intervention get 2 weeks of free public transport during summer if they order it. The authors would like to know if a nudge or an incentive work best to get people to use public transport and to keep people using the public transport after the summer ends.
Nudge: Social norm, tell people how many other people in Skåne are taking public transport. The given nudge, “Total reser 72% av oss skåningar kollektiv någon gang. “
Economic incentive: Give people two more weeks of free travel. The given incentive, “Erbjudanted är värt 1250 SEK”.
Study: An randomized controlled trial (RCT - på dansk lodtrækningsstudie) with Skånetrafik.
Control group - 4737 cards with 2 weeks of free public transport
Social norm group - 4737 cards with 2 weeks of free public transport plus the social norm information.
Incentive group - 4737 cards with 2 weeks of free public transport plus 2 extra weeks free.
Outcomes we are looking at -
Number of cards ordered
Number of cards activated
Number of cards used 6 months later
Results
We see that in the case of the nudge, we do not really have the big differences from the control group. We see a slightly higher procentage of people who activated the card and who used it after summer.
But in the case of the economic incentive we have a relative big increase in people who got the card, activated it and used it after the summer. This could mean, that using economic incentives to get people to use public transport actually works better than introducing social norm nudge.
Tailor:
Economic incentives are more effectful to change behavior in this context.
