Verkorte samenvatting Human Decision Making
General decision making - Introduction
Rational decision making
Economists think people make rational decisions. The rational decision making process looks like this:
1. Gather information
2. Evaluation
3. Action: deciding/choosing
4. Implementation: executing the decision
5. Evaluation of the outcome
Judgement vs problem solving:
Judgement: cognitive part of decision making, part that comes before making a decision.
Picking one of the options
Problem solving: looks like decision making, only it not necessarily requires action
Nowadays we know that rational decision making is not possible in a lot of situations and that
behavioural economics studies deviations from this rational model.
At least: economics think it can be seen as deviations from a model and still be the same
theory
o Psychology makes new theories about the deviations from a model
Bounded rationality
For the rest of the course, people focus on bounded rationality: people don’t have all the
information to be rational and therefore use simplified models and heuristics to make a choice
Intertemporal decision making: making decisions over time. This is present in almost every domain
in life. For example eating behaviour: choosing in the supermarket what you will eat in a week.
Discounting
Discounting: the valuation of future outcomes relative to present outcomes.
In other words: the weight you give to different outcomes depending on when they will come to
effect. For example: how much would are you willing to give to receive something now instead of in a
week?
Measured with discount rate (dr), where a higher rate indicates more impatience.
Models of discounting
Exponential discounting: way you should discount according to economic rational model.
Constant discount rate over time
So if the amount of time doubles, the discount rate doubles too.
Hyperbolic discounting: most realistic model, used by people a lot.
High discount rate in near future, lower discount rate in the distant future
Leads to:
o Preference for smaller-sooner (SS) over larger-later (LL)
People prefer 50 euros now over 100 euros in a year
o Dynamic inconsistency: choices that you make depend on the length of the delay
there is until these choices come into effect.
o Preference reversal: now you prefer A, later you might prefer B
, Deviation
Hyperbolic discounting is a way in which people deviate from the standard way of discounting. This
leads to:
Suboptimal outcomes (according to rational model)
Satisfactory outcomes (according to reality)
Types of deviating
Delay effect: lower discount rates for longer delays.
o Study: 15 dollars now or..?
20 in a month (dr 345%)
100 in 10 years (dr 19%)
Magnitude effect: higher discount rates for smaller amounts
o Study:
15 dollars now equals 30 in 3 months (dr 277%)
3000 dollars now equals 3500 in 3 months (dr 62%)
Sign effect: higher discount rates for gains than for losses
o Remember 15 dollars now was 30 in 3 months (dr 277%)
o But if people got a fine (15 dollars), which they had to pay now or in a month, they
would want to pay only 1 euro extra (dr 26%)
Direction effect: higher rates for increased delays dan decreased delays
o Study: 100 dollar gift card
Delay premium: Now or 6 months, people wanted 26 dollars extra for 6
months
Speed-up premium: 6 months or now, people willing to pay 10 dollars for
now
Sequence effect: increased sequences are preferred over decreased sequences. Increasing or
decreasing salary? End salary is the same in both options. People prefer increasing.
o Savouring vs dread
Willingness to delay a pleasurable outcome
Willingness to speed-up a painful outcome
Interval effect: higher discount rates for shorter intervals
o People give higher discount rates to three interval periods of 8 months than 1 period
of 24 months
Between these effects, there is a lot of interaction
Measuring discount rates
Almost all of the effects above are time trade-off measures
o Giving options between smaller sooner and larger later rewards
o Option to do it incentivized or non-incentivized: get the money you choose or
hypothetical situation
Other way to measure: inferring instead of measuring. The researcher infers instead of really
measuring it.
o Can be done from behaviour: someone buying a cheap energy consuming dryer of an
expensive energy sustainable. Both cost the same in the end, but a researcher can
infer from this choice what a person wants when discounting
General decision making - Introduction
Rational decision making
Economists think people make rational decisions. The rational decision making process looks like this:
1. Gather information
2. Evaluation
3. Action: deciding/choosing
4. Implementation: executing the decision
5. Evaluation of the outcome
Judgement vs problem solving:
Judgement: cognitive part of decision making, part that comes before making a decision.
Picking one of the options
Problem solving: looks like decision making, only it not necessarily requires action
Nowadays we know that rational decision making is not possible in a lot of situations and that
behavioural economics studies deviations from this rational model.
At least: economics think it can be seen as deviations from a model and still be the same
theory
o Psychology makes new theories about the deviations from a model
Bounded rationality
For the rest of the course, people focus on bounded rationality: people don’t have all the
information to be rational and therefore use simplified models and heuristics to make a choice
Intertemporal decision making: making decisions over time. This is present in almost every domain
in life. For example eating behaviour: choosing in the supermarket what you will eat in a week.
Discounting
Discounting: the valuation of future outcomes relative to present outcomes.
In other words: the weight you give to different outcomes depending on when they will come to
effect. For example: how much would are you willing to give to receive something now instead of in a
week?
Measured with discount rate (dr), where a higher rate indicates more impatience.
Models of discounting
Exponential discounting: way you should discount according to economic rational model.
Constant discount rate over time
So if the amount of time doubles, the discount rate doubles too.
Hyperbolic discounting: most realistic model, used by people a lot.
High discount rate in near future, lower discount rate in the distant future
Leads to:
o Preference for smaller-sooner (SS) over larger-later (LL)
People prefer 50 euros now over 100 euros in a year
o Dynamic inconsistency: choices that you make depend on the length of the delay
there is until these choices come into effect.
o Preference reversal: now you prefer A, later you might prefer B
, Deviation
Hyperbolic discounting is a way in which people deviate from the standard way of discounting. This
leads to:
Suboptimal outcomes (according to rational model)
Satisfactory outcomes (according to reality)
Types of deviating
Delay effect: lower discount rates for longer delays.
o Study: 15 dollars now or..?
20 in a month (dr 345%)
100 in 10 years (dr 19%)
Magnitude effect: higher discount rates for smaller amounts
o Study:
15 dollars now equals 30 in 3 months (dr 277%)
3000 dollars now equals 3500 in 3 months (dr 62%)
Sign effect: higher discount rates for gains than for losses
o Remember 15 dollars now was 30 in 3 months (dr 277%)
o But if people got a fine (15 dollars), which they had to pay now or in a month, they
would want to pay only 1 euro extra (dr 26%)
Direction effect: higher rates for increased delays dan decreased delays
o Study: 100 dollar gift card
Delay premium: Now or 6 months, people wanted 26 dollars extra for 6
months
Speed-up premium: 6 months or now, people willing to pay 10 dollars for
now
Sequence effect: increased sequences are preferred over decreased sequences. Increasing or
decreasing salary? End salary is the same in both options. People prefer increasing.
o Savouring vs dread
Willingness to delay a pleasurable outcome
Willingness to speed-up a painful outcome
Interval effect: higher discount rates for shorter intervals
o People give higher discount rates to three interval periods of 8 months than 1 period
of 24 months
Between these effects, there is a lot of interaction
Measuring discount rates
Almost all of the effects above are time trade-off measures
o Giving options between smaller sooner and larger later rewards
o Option to do it incentivized or non-incentivized: get the money you choose or
hypothetical situation
Other way to measure: inferring instead of measuring. The researcher infers instead of really
measuring it.
o Can be done from behaviour: someone buying a cheap energy consuming dryer of an
expensive energy sustainable. Both cost the same in the end, but a researcher can
infer from this choice what a person wants when discounting
