HC 1:
Nudge: any aspect of the ‘choice architecture’ that alters people’s behavior in a predictable way
without forbidding options or dramatically changing economic incentives. They are not mandates,
they must be cheap and easy to avoid. They must also be transparent.
Traditional economics gets people to do things through monetary incentives. Nudging is an
alternative way, that follows behavioral economics to get people to do things. However the standard
economic model of consumer behavior assumes the following things:
People are rational
People act with full information (full external knowledge)
People have known preferences (full internal knowledge)
People choose the best option available (they maximize utility)
Liberalism states that the state should have respect for all citizens as free and equal human beings, it
should enable all citizens to develop and pursue their own conception of the good life and harm to
others is the only basis for legitimate government intervention.
Paternalism goes further than the minimalist ‘no harm’ principle of liberalism and argues that people
should also be protected against harming themselves. The state should promote the interests of all
citizens in living secure, healthy, wealthy and happy lives and it should interfere with people’s
liberties if this generates desirable outcomes.
Libertarian paternalism or ‘nudging’ aims to improve people’s choices by being both:
Libertarian: it does not block people’s choices and thus respects people’s (‘negative’ or
‘formal’) liberty.
Paternalistic: it makes people better off and thus improves their well-being (health, wealth
and happiness).
The principles of nudging are as follows:
Psychological and affective ways of changing behavior
Focus on “guiding” behavior (instead of influencing)
Instead of using incentives and information, use cues, frames, defaults.
No force or prohibition! Preserve freedom of choice
Minimal costs
Change of choice architecture
Choice architecture is the changing of the environment in which the choice is being made (choice
environment) so as to encourage the chooser to select a preferred choice. It maintains the chooser’s
freedom to select other choices, it simply makes both the desirable- and undesirable choice
available, only the undesirable choice harder.
Nudging can help especially in the following situations that create problems:
Benefits now – costs later (should-want conflicts)
Problems with a low frequency
Choices without feedback
Very difficult problems (e.g. choosing the specific type of mortgage)
Outcomes of choices difficult to predict
HC 2: intuition, overconfidence and defaults
,Behavioral economics assumes humans predictably deviate from “optimality”, for example:
Heuristics: mental “shortcuts” that sometimes backfire.
Overreaction to losses: people detest losing.
Status quo: stick to what we have (inertia).
Social norms: other-regarding preferences, fairness concerns and herd mentality.
Impulsivity: we want rewards right now.
Optimism and overconfidence: help us maintain positive self-concept.
Intuition: expressed judgments retain hypothesized initial proposal with little modification.
Intuition trap: intuition is often wrong
Confirmation bias: we are resistant to changing our prior beliefs and we focus on confirming
instances and dismiss disconfirming ones.
Your preferences are not representative, which leads to: projection bias/false consensus
Overconfidence: person’s subjective confidence is greater than the objective accuracy of those
judgments. It happens due to option fixation which is that people tend to focus on their first guess,
while ignoring alternatives. It happens on hard rather than easy tasks.
Planning fallacy: people underestimate how long it will take them to get things done. This happens
especially for long and complicated tasks, people underestimate their own completion times but not
others and especially when motivation to complete the task increases. This happens because we
often focus on the future while ignoring related past events, we overlook important potential
obstacles and the fallacy can be eliminated when connecting events with past experiences.
Optimism bias: a cognitive illusion which leads to the tendency to overestimate our likelihood of
experiencing good events in our lives and underestimate our likelihood of experiencing bad events.
When given information on actual percentage of experiencing a certain bad event people would only
update their predictions if the information was positive (e.g. if 50% 35%, if 1% 11%). Nudging
against optimism bias can be done through giving feedback and asking people to explain why they
think their answer is correct.
Dunning-kruger effect: the overestimation of own skills or knowledge, especially those who are
incompetent. This leads to a double burden, people make wrong decisions and are unaware of it.
they thus miss metacognition (less able to recognize competence), are unable to use information to
adjust judgments, but training helps: incompetent people then realize that they do badly. Thus if
people don’t know what they don’t know, they are confident that they know a lot. When people
know what they don’t know they are less confident about their knowledge. Incompetence causes
poor performance and the inability to recognize that one’s performance is poor (lack of
metacognitive skills).
, Defaults framework of Goldstein:
Mass defaults: apply to all customers
Benign defaults: best guess, most acceptable to most customers (one-size-fits all).
Random defaults: arbitrarily assigned to one of the default options.
Hidden options: default is only choice, but hard to find alternatives exist.
Personalized defaults: tailored to individual needs
Smart defaults: Set to specific segment or individual based on demographic or geographic
variables.
Persistent defaults: Based on former choice (higher satisfaction and loyalty, but annoyance
when preferences change).
Adaptive defaults: Update over time depending on information given.
The reasons that defaults work for example in donor programs are: loss aversion (avoid potential
losses generated by change), no effort (changing is effortful), default is perceived to be the
recommended action by policy makers/employer and preselected option receives more attention.
Possible outcomes of setting defaults (for example in donating to charity):
Direct default effect: increases the choice of set default.
Scale-back effect of low defaults: People who otherwise donated a larger amount will now
donate the default. Anchoring: even if the low default is not chosen, people will still scale
back. No participation: low default => charity needs no donations
Backfire effect of high defaults: people might see the default as a persuasive attempt, which
could lead to a reduction of donation rates.
Lower bar effect of low defaults: Increase of participation, as people feel less bad by
contributing only what they can afford.
However evidence found that there isn’t evidence for the optimistic prediction (direct default effect)
and also not for the pessimistic prediction (backfire effect). There was evidence for two novel
effects: Scale-back effect for low defaults, but counteracted by lower-bar effect.
Nudge: any aspect of the ‘choice architecture’ that alters people’s behavior in a predictable way
without forbidding options or dramatically changing economic incentives. They are not mandates,
they must be cheap and easy to avoid. They must also be transparent.
Traditional economics gets people to do things through monetary incentives. Nudging is an
alternative way, that follows behavioral economics to get people to do things. However the standard
economic model of consumer behavior assumes the following things:
People are rational
People act with full information (full external knowledge)
People have known preferences (full internal knowledge)
People choose the best option available (they maximize utility)
Liberalism states that the state should have respect for all citizens as free and equal human beings, it
should enable all citizens to develop and pursue their own conception of the good life and harm to
others is the only basis for legitimate government intervention.
Paternalism goes further than the minimalist ‘no harm’ principle of liberalism and argues that people
should also be protected against harming themselves. The state should promote the interests of all
citizens in living secure, healthy, wealthy and happy lives and it should interfere with people’s
liberties if this generates desirable outcomes.
Libertarian paternalism or ‘nudging’ aims to improve people’s choices by being both:
Libertarian: it does not block people’s choices and thus respects people’s (‘negative’ or
‘formal’) liberty.
Paternalistic: it makes people better off and thus improves their well-being (health, wealth
and happiness).
The principles of nudging are as follows:
Psychological and affective ways of changing behavior
Focus on “guiding” behavior (instead of influencing)
Instead of using incentives and information, use cues, frames, defaults.
No force or prohibition! Preserve freedom of choice
Minimal costs
Change of choice architecture
Choice architecture is the changing of the environment in which the choice is being made (choice
environment) so as to encourage the chooser to select a preferred choice. It maintains the chooser’s
freedom to select other choices, it simply makes both the desirable- and undesirable choice
available, only the undesirable choice harder.
Nudging can help especially in the following situations that create problems:
Benefits now – costs later (should-want conflicts)
Problems with a low frequency
Choices without feedback
Very difficult problems (e.g. choosing the specific type of mortgage)
Outcomes of choices difficult to predict
HC 2: intuition, overconfidence and defaults
,Behavioral economics assumes humans predictably deviate from “optimality”, for example:
Heuristics: mental “shortcuts” that sometimes backfire.
Overreaction to losses: people detest losing.
Status quo: stick to what we have (inertia).
Social norms: other-regarding preferences, fairness concerns and herd mentality.
Impulsivity: we want rewards right now.
Optimism and overconfidence: help us maintain positive self-concept.
Intuition: expressed judgments retain hypothesized initial proposal with little modification.
Intuition trap: intuition is often wrong
Confirmation bias: we are resistant to changing our prior beliefs and we focus on confirming
instances and dismiss disconfirming ones.
Your preferences are not representative, which leads to: projection bias/false consensus
Overconfidence: person’s subjective confidence is greater than the objective accuracy of those
judgments. It happens due to option fixation which is that people tend to focus on their first guess,
while ignoring alternatives. It happens on hard rather than easy tasks.
Planning fallacy: people underestimate how long it will take them to get things done. This happens
especially for long and complicated tasks, people underestimate their own completion times but not
others and especially when motivation to complete the task increases. This happens because we
often focus on the future while ignoring related past events, we overlook important potential
obstacles and the fallacy can be eliminated when connecting events with past experiences.
Optimism bias: a cognitive illusion which leads to the tendency to overestimate our likelihood of
experiencing good events in our lives and underestimate our likelihood of experiencing bad events.
When given information on actual percentage of experiencing a certain bad event people would only
update their predictions if the information was positive (e.g. if 50% 35%, if 1% 11%). Nudging
against optimism bias can be done through giving feedback and asking people to explain why they
think their answer is correct.
Dunning-kruger effect: the overestimation of own skills or knowledge, especially those who are
incompetent. This leads to a double burden, people make wrong decisions and are unaware of it.
they thus miss metacognition (less able to recognize competence), are unable to use information to
adjust judgments, but training helps: incompetent people then realize that they do badly. Thus if
people don’t know what they don’t know, they are confident that they know a lot. When people
know what they don’t know they are less confident about their knowledge. Incompetence causes
poor performance and the inability to recognize that one’s performance is poor (lack of
metacognitive skills).
, Defaults framework of Goldstein:
Mass defaults: apply to all customers
Benign defaults: best guess, most acceptable to most customers (one-size-fits all).
Random defaults: arbitrarily assigned to one of the default options.
Hidden options: default is only choice, but hard to find alternatives exist.
Personalized defaults: tailored to individual needs
Smart defaults: Set to specific segment or individual based on demographic or geographic
variables.
Persistent defaults: Based on former choice (higher satisfaction and loyalty, but annoyance
when preferences change).
Adaptive defaults: Update over time depending on information given.
The reasons that defaults work for example in donor programs are: loss aversion (avoid potential
losses generated by change), no effort (changing is effortful), default is perceived to be the
recommended action by policy makers/employer and preselected option receives more attention.
Possible outcomes of setting defaults (for example in donating to charity):
Direct default effect: increases the choice of set default.
Scale-back effect of low defaults: People who otherwise donated a larger amount will now
donate the default. Anchoring: even if the low default is not chosen, people will still scale
back. No participation: low default => charity needs no donations
Backfire effect of high defaults: people might see the default as a persuasive attempt, which
could lead to a reduction of donation rates.
Lower bar effect of low defaults: Increase of participation, as people feel less bad by
contributing only what they can afford.
However evidence found that there isn’t evidence for the optimistic prediction (direct default effect)
and also not for the pessimistic prediction (backfire effect). There was evidence for two novel
effects: Scale-back effect for low defaults, but counteracted by lower-bar effect.
