Lecture 18 Nudges

  • What is a nudge?
    • aspect of CA that alters people's behavior in a predictable way
    • without forbidding any options or without significantly changing their economic incentives
    • The intervention should be easy and cheap to avoid
    • Something that changes the incentives of Homo sapiens (ECONs) can(not) be called a nudge
  • Examples
    • 2009 Piano Stair in Odenplan Metro Station, Sweden.
      • Novel ideas in induce people to use stairs, rather than escalators
      • Encouraged 66% more people to use stairs
    • "Eat Healthy" by Google
      • Candy in Opaque bins
      • Salad in full view, dessert much further down
      • "People with bigger plates eat more"
      • color-coding foods by healthiness
    • Alternate versions of Job interviews to unemployed individuals
      1. Without "Hi Sam" - 11% turned up
      2. With "Hi Sam" - 15%
      3. With "Hi Sam" and end with "Good luck, Michael" - 27%
    • Fitness and similar apps: "You are good! Continue doing it"
    • Nudges can be beneficial for Nudgers as well.
      • Black housefly image into each urinal
      • Reduced spillage by 80%
      • Benefitted the nudgers (Janitors)
  • Priming
    • Automatic System's mysterious working
    • Hint (idea/concept) \(\implies\) trigger association \(\implies\) stimulate action
    • Social scientists have discovered that when they measure people's intentions, they affect their conduct
      • "Are you planning to purchase?" - "Yes" - and some % of them actually stick to their commitment
    • "mere-measurement effect": when asked what they intend to do, they become more likely to act the same.
    • Effective by offering simple and apparently irrelevant cues
      • e.g. Briefcases and boardroom tables, make people more competitive, less cooperative and less generous
  • Properties of Nudges
    • libertarian paternalism: make better decisions for yourselves, with little to no cost to them, without affecting the choices of those who are already rational and well informed
    • And for those not, it is potentially beneficial for them, by their own lights
  • Libertarian Paternalism
    • Libertarian aspect: straightforward insistence "You are free to do what you like to, or opt out if you find it undesirable"
    • Paternalistic aspect: influence behavior in such a way \(\implies\) live longer, healthier and better.
  • Why need a Nudge?
    1. Fraught Choices - Self-control challenges
      • problems arise when making decision that test people's self-control (mostly when choices and their consequences are deferred)
      • We can call it
        • investment goods: exercise, flossing, dieting.
          • People err on doing too little.
        • sinful goods: smoking, alcohol, chocolate doughnuts
          • Pleasure now, suffer later
      • both types are prime candidates for nudges
    2. Degree of difficulty
      • Hard problems are good for nudges
      • Picking the right mortgage vs choosing the right loaf of bread
    3. Frequency
      • Hard problems + practice = Easier problems
      • Unfortunately, most important decisions are not frequent, no can practice
      • higher the stakes \(\implies\) less often we are able to practice (practiced at grocery shopping but not buying houses)
    4. Feedback
      • But what if no learning opportunities?
        • Golfing without being able to see where the ball ends up (just like many of life's choices)
        • High fat diet for years without any warnings signs until the heart attack
      • When feedback doesn't work, nudge helps
    5. When preferences are uncertain and uninformed
      • Smart tourists rely on others (e.g. waiters) for help
      • hard to decide when trouble translating choices they face into experiences they have
      • When hard time predicting outcomes of their choices on life:
        • they have less to gain by numerosity of options and choosing on their own
  • How successful or appropriate are nudges?
    • What may work in one situation with a group may not work in another (situation and/or group)
    • UK House of Lords Science & Tech committee 2011 report: concert about ethical issues because nudges alter behavior through mechanisms people aren't aware.
    • Preference purification - Nudges are to nudge HUMANS towards ECONs (it is used to purify the contaminated revealed preferences)
    • Sudgen claims that, given a truly free choice, a SuperReasoner1 may not choose the best option since their processing ability is objective. But preferences are subjective
    • Though, if successfully nudged, the SuperReasoner
      • is worse off according to his judgement
      • may not be according to the CA's judgement
  • Nudge as a policy tool
    • Four good reasons to nudge someone
      • facing a very complicated choice
      • having had very little experience
      • being relatively uninformed
      • having self-control problems
  • Effective Use of Nudges
    • Framework by Thaler and Sunstein (2008)
      1. iNcentives - people understand nudges only if they are salient (e.g. price and cost)
      2. Understand - mapping choices to outcomes, people need help doing so
      3. Defaults - matter a lot because of present bias and choice overload**
      4. Give feedback - People learn, so tell them if they are doing it right or wrong
      5. Expect error - Need forgiving systems
      6. Structure complex choices - Keep things simple to avoid overload and errors
    • Not at the sacrifice of liberty, without forbidding any options or significantly changing economic incentives. Easy and cheap to avoid intervention
    • Don't change the set of options. Change the option that will be chosen
    • Lack of neutral option is key here.
    • Use design systems to make people do the things that they want to do but find it hard to do for behavioral reasons.
    • UK Policy-makers use MINDSPACE
  • MINDSPACE #later
  • BASIC - Method to design nudges #later

  1. Access to all relevant info, no cognitive limitation, perfect self-control, a real ECON ↩