Hoorcollege 7 Fairness
The Big Picture
● You learned that markets are often able to allocate resources in an efficient way:
market failures
● Public goods
○ Externalities & internalities
○ Imperfect competition
● But you also know about market failures, i.e. situations in which markets cannot
deliver efficient outcomes
○ market failures and government intervention as a reaction
● Is efficiency the only criteria to judge market outcomes?
● What about fairness?
○ Fairness is a criteria that shows up in many levels in a society
○ Equal and fair treatments are principals that almost everyone adheres
○ Can we include fairness in our analysis of markets?
Today’s questions
● Are efficiency and fairness related to each other?
● Can we define fairness precisely?
○ Much more complex to define
● Can we compare the outcomes of two persons?
○ We need to compare how person A and person B feel about an outcome
● Can we measure people’s ideas about fairness?
● How can we implement outcomes that are fair for society?
○ Practical applications of the concepts
● Refresher: Pareto efficiency
● Fairness
- The relation between fairness and efficiency
- What people think about fairness
- The theory of fair allocations
- Social welfare functions –covered in the book
● Conclusions & Take away msg
Pareto efficiency
● We say that a societal outcome x is a pareto improvement on an outcome y if
○ At least one person strictly prefers x to y, (at least one person is better off)
○ All persons weakly prefer x to y (no one is hurt)
● When we make a pareto improvement, nobody is worse off, hence, nobody objects
against making pareto improvements
● A societal outcome x is pareto efficient if there does not exist a pareto improvement
of it
○ We cannot pareto improve x
● In short, we cannot do better at all
,Pareto efficiency
● Economics are advocates of efficiency and have been criticised for only looking at
pareto efficiency
○ Fairness is intuitively a notion that we want to take into account when judging
market outcomes
○ In everyday language, efficiency is used for other meanings as we discussed
earlier:
■ Cost savings even if this means achieving less
■ Cost shifting: we reduce some cost, by increasing others
● Terminology needs to be clear: pareto efficiency means that it is impossible to obtain
an outcome that is better for everybody
● Who can be against efficiency?
Imagine
● A world with only bread
● Fixed amount of each good, shared between adam, eve
Relationship between pareto efficiency & fairness
● Assume you are the social planner (government) and your society constitutes of a
small population: adam and eve
● You have 10 breads to divide between adam and eve, and that bread is a standard
good for both (nobody dislikes having more of it)
○ There are two allocation you can implement (both of which are efficient)
○ Allocation A = adam 1 bread, eve 9 breads
○ Allocation B = adam 5 breads, eve 5 breads
, ● Why are both pareto improvements?
○ One cannot make one agent better off without hurting the other
○ Underlying assumption: they both like bread
● Now imagine that adam and eve are equally deserving
○ They worked equal amounts
● How can we select the fairest allocation?
○ One cannot make one agent better off without hurting the other
○ What does fair mean?
● How can we select the fairest allocation?
○ Equal share?
■ Underlying assumption: they have equal marginal utility (they like
bread to the same extend)
● Is this a reasonable assumption?
● This is a stronger assumption than saying we have an efficient
allocation (where we say they both like bread)
■ How can we measure that?
● What if adam likes bread more than eve?
● Ask them how much they like bread in a scale of 0-10
● Problem of comparability: but what if they use different scales?
How can we interpret their answers?
● Potential solution for policy makers:
○ We cannot decide how much they like bread if we just ask them
○ Work with monetary amounts instead:
○ Give bread a monetary value, look at market
○ And make a reasonable assumption that everyone has the same utility
function for money
○ If 1 bread equals 1 euros, and they are equally deserving
Take away msg: we need to compare utilities to talk about fairness and it is not obvious how
to do this
Pareto efficiency & fairness
● Expressing everything in monetary terms is practical
○ Even for things that do not have a market price
○ We can elicit people’s willingness to pay (WTP)
● To make a decision as a policy maker, we always face a trade-off:
○ Benefits
○ Costs
● Example: during corona how much would you be willing to pay to go back to
campus? As a manager, understand whether it is worth to come back to campus?
○ Benefits: students learn better (how much students value this, elicit their
WTP)
○ Costs: safety measures
, ● Some things are harder to evaluate. Example related to health care costs:
○ Do you prefer to live healthy for a short time or with a disease for a longer
time?
● Quality-adjusted life year (QALY): quantifies benefits of treatment - lenght of life and
quality of life - into a single number. This can be used for making cost-effectiveness
analysis
○ Ways to quantify the benefits and costs of treatments
■ Quality of life (questions like: like a healthy life of 10 years, and live for
15 years without a leg)
■ Costs of treatments
To sum up:
● Efficiency is easy to
○ Define
○ To agree with
● We care about fairness as much as we care about efficiency
○ Want societal outcomes to be fair
● However, even when two people deserve an equal treatment (so fairness it is easy to
define) we may need to make interpersonal comparisons
● Making interpersonal comparisons require
○ Strong assumptions
○ Difficult measures
Why fairness is left out:
● Fairness is difficult to formalize: even academics (economists, psychologist,
philosophers, etc..) do not agree among each other.
● Fairness is often overused and used in a very instrumental way in the political
discourse (ideological positions)
● Politicians often appeal to fairness but the notion of fairness to which they appeal is
not clearly defined & used in a self-serving way:
○ “Taxes are about more than money and they’re about more than economics.
They’re about fairness, and this bill is fair” (Senator Robert Packwood)
○ “For our economy, this is the wrong bill at the wrong time ... making deficit
reduction more difficult and less fair”(SenatorCarl M. Levin)
● Some forms of fairness (such as full equality) may lead to less efficiency
● Implementing fair outcomes may distort market incentives and lead to some welfare
losses (within classical theory)
Fairness - empirical evidence
● Recently economists started to be interested in fairness
○ To study the fairness perceptions across countries to help us understand why
some countries have large welfare states and some others have not
○ Use of new techniques: conduction empirical research - surveys, experiments
- on people’s fairness views
■ Experimental economics: new
● There is a renewed interest in fairness and the purpose is to
○ Learn what people consider to be fair (let the data speak)
The Big Picture
● You learned that markets are often able to allocate resources in an efficient way:
market failures
● Public goods
○ Externalities & internalities
○ Imperfect competition
● But you also know about market failures, i.e. situations in which markets cannot
deliver efficient outcomes
○ market failures and government intervention as a reaction
● Is efficiency the only criteria to judge market outcomes?
● What about fairness?
○ Fairness is a criteria that shows up in many levels in a society
○ Equal and fair treatments are principals that almost everyone adheres
○ Can we include fairness in our analysis of markets?
Today’s questions
● Are efficiency and fairness related to each other?
● Can we define fairness precisely?
○ Much more complex to define
● Can we compare the outcomes of two persons?
○ We need to compare how person A and person B feel about an outcome
● Can we measure people’s ideas about fairness?
● How can we implement outcomes that are fair for society?
○ Practical applications of the concepts
● Refresher: Pareto efficiency
● Fairness
- The relation between fairness and efficiency
- What people think about fairness
- The theory of fair allocations
- Social welfare functions –covered in the book
● Conclusions & Take away msg
Pareto efficiency
● We say that a societal outcome x is a pareto improvement on an outcome y if
○ At least one person strictly prefers x to y, (at least one person is better off)
○ All persons weakly prefer x to y (no one is hurt)
● When we make a pareto improvement, nobody is worse off, hence, nobody objects
against making pareto improvements
● A societal outcome x is pareto efficient if there does not exist a pareto improvement
of it
○ We cannot pareto improve x
● In short, we cannot do better at all
,Pareto efficiency
● Economics are advocates of efficiency and have been criticised for only looking at
pareto efficiency
○ Fairness is intuitively a notion that we want to take into account when judging
market outcomes
○ In everyday language, efficiency is used for other meanings as we discussed
earlier:
■ Cost savings even if this means achieving less
■ Cost shifting: we reduce some cost, by increasing others
● Terminology needs to be clear: pareto efficiency means that it is impossible to obtain
an outcome that is better for everybody
● Who can be against efficiency?
Imagine
● A world with only bread
● Fixed amount of each good, shared between adam, eve
Relationship between pareto efficiency & fairness
● Assume you are the social planner (government) and your society constitutes of a
small population: adam and eve
● You have 10 breads to divide between adam and eve, and that bread is a standard
good for both (nobody dislikes having more of it)
○ There are two allocation you can implement (both of which are efficient)
○ Allocation A = adam 1 bread, eve 9 breads
○ Allocation B = adam 5 breads, eve 5 breads
, ● Why are both pareto improvements?
○ One cannot make one agent better off without hurting the other
○ Underlying assumption: they both like bread
● Now imagine that adam and eve are equally deserving
○ They worked equal amounts
● How can we select the fairest allocation?
○ One cannot make one agent better off without hurting the other
○ What does fair mean?
● How can we select the fairest allocation?
○ Equal share?
■ Underlying assumption: they have equal marginal utility (they like
bread to the same extend)
● Is this a reasonable assumption?
● This is a stronger assumption than saying we have an efficient
allocation (where we say they both like bread)
■ How can we measure that?
● What if adam likes bread more than eve?
● Ask them how much they like bread in a scale of 0-10
● Problem of comparability: but what if they use different scales?
How can we interpret their answers?
● Potential solution for policy makers:
○ We cannot decide how much they like bread if we just ask them
○ Work with monetary amounts instead:
○ Give bread a monetary value, look at market
○ And make a reasonable assumption that everyone has the same utility
function for money
○ If 1 bread equals 1 euros, and they are equally deserving
Take away msg: we need to compare utilities to talk about fairness and it is not obvious how
to do this
Pareto efficiency & fairness
● Expressing everything in monetary terms is practical
○ Even for things that do not have a market price
○ We can elicit people’s willingness to pay (WTP)
● To make a decision as a policy maker, we always face a trade-off:
○ Benefits
○ Costs
● Example: during corona how much would you be willing to pay to go back to
campus? As a manager, understand whether it is worth to come back to campus?
○ Benefits: students learn better (how much students value this, elicit their
WTP)
○ Costs: safety measures
, ● Some things are harder to evaluate. Example related to health care costs:
○ Do you prefer to live healthy for a short time or with a disease for a longer
time?
● Quality-adjusted life year (QALY): quantifies benefits of treatment - lenght of life and
quality of life - into a single number. This can be used for making cost-effectiveness
analysis
○ Ways to quantify the benefits and costs of treatments
■ Quality of life (questions like: like a healthy life of 10 years, and live for
15 years without a leg)
■ Costs of treatments
To sum up:
● Efficiency is easy to
○ Define
○ To agree with
● We care about fairness as much as we care about efficiency
○ Want societal outcomes to be fair
● However, even when two people deserve an equal treatment (so fairness it is easy to
define) we may need to make interpersonal comparisons
● Making interpersonal comparisons require
○ Strong assumptions
○ Difficult measures
Why fairness is left out:
● Fairness is difficult to formalize: even academics (economists, psychologist,
philosophers, etc..) do not agree among each other.
● Fairness is often overused and used in a very instrumental way in the political
discourse (ideological positions)
● Politicians often appeal to fairness but the notion of fairness to which they appeal is
not clearly defined & used in a self-serving way:
○ “Taxes are about more than money and they’re about more than economics.
They’re about fairness, and this bill is fair” (Senator Robert Packwood)
○ “For our economy, this is the wrong bill at the wrong time ... making deficit
reduction more difficult and less fair”(SenatorCarl M. Levin)
● Some forms of fairness (such as full equality) may lead to less efficiency
● Implementing fair outcomes may distort market incentives and lead to some welfare
losses (within classical theory)
Fairness - empirical evidence
● Recently economists started to be interested in fairness
○ To study the fairness perceptions across countries to help us understand why
some countries have large welfare states and some others have not
○ Use of new techniques: conduction empirical research - surveys, experiments
- on people’s fairness views
■ Experimental economics: new
● There is a renewed interest in fairness and the purpose is to
○ Learn what people consider to be fair (let the data speak)
