EC201 Micro LT

Question 1

The Market System

Answer 1

• Decentralisation- people are being instructed or explicitly coordinated
• Greed - they do not care about people that they are producing for/trading with
• What we mean by this is a way of producing goods allocating resources to people using price signals
• We refer to a “system” because we are looking at the whole economy and how it works rather than any specific sector or market
• There are many grand debates about the merits of markets and what they achieve:
forces us to think about what we value in human welfare
efficiency, equality, freedom etc.
• We want our models to give some insight into these issues

Question 2

What is a good?

Answer 2

• Goods may not just be things that we conventional recognise as goods: mobile phone services, food, paper, pens etc.
• A good can be different according to where and when it is consumed so they are distinct goods and prices could vary if supply and demand for the good need to be equalised on different dates
• There are complete markets when all possible variants of goods can be traded

Question 3

Market Goods

Answer 3

• The focus is on those things that are traded at defined prices
• Statistical agencies collect price data in markets and data on what people buy and produce
• These are used to get measures of aggregate activity in the market economy (important for guiding policy)
• The non-market sector is also important: home production, government production, some parts of the non-profit sector
• Demand for each good has standard properties: goods can be complements and substitutes, but each individual only supplies labour in the area in which they have a talent

Question 4

General Equilibrium

Answer 4

• Prices are set so as to “clear” all markets simultaneously
• Consumers are able to get what they want given these prices
• Given the distribution of talents, the market delivers enough goods to provide for what consumers want
• We will denote equilibrium prices as p* = (p1,… pN)
• The best way to think about this is in terms of “excess demand”, where excess demand = demand - supply
• When markets clear, all excess demand are zero

Question 5

Intuitions about Excess Demand Functions

Answer 5

• More people who are able to produce a particular good
• Then excess demand will fall for any given price
• We would then expect the market clearing price to fall
• But it will also affect markets for complements and substitutes for that good

Question 6

If a good becomes more popular

Answer 6

• Then excess demand for that good increases
• We would expect the market clearing price to rise
• Will affect other markets which are complements or substitutes

Question 7

Price normalisation

Answer 7

• A general equilibrium can be expressed entirely in terms of numeraire good since only relative prices matter
• But the choice of numeraire is arbitrary: we can set pk = 1 for any k

Question 8

Walras Law

Answer 8

• If all but one market is in equilibrium, then so is the last one
• To see this, set p1 = 1 and note that demand for X1 + sum of demand for all other X1s = Demand for Z1 + sum of demand for all other Z1s

Question 9

Consistency

Answer 9

• Main breakthroughs came in the second half of the 20th century
• Often called an Arrow-Debreu economy
• Key innovation was to apply “fixed point” reasoning to the market economy

Question 10

Excess Demand in a Single Market

Answer 10

• High enough price discourages demand and encourage supply
• Low enough price encourages demand and discourages supply
• No jumps in supply and demand
• Importance of convexity assumptions to make supply and demand continuous
• Averaging over people also helps too

Question 11

Extension to Multiple Markets

Answer 11

• Cannot be drawn by the basic idea is the same as a single market except that all prices are moving to achieve equilibrium across all markets
• However, continuity and what happens in a market as prices become high and low still play a key role
• A set of prices that set all excess demand functions equal to zero simultaneously is known as a “fixed point”
• A market system in this very stylised world is set of market-clearing prices

Question 12

The Role of Government

Answer 12

• The general equilibrium model is generally though of as a “pure market system”
• But this is misleading
• Two key things that government might do: enforce property rights, i.e. ensure that people own what they produce, enforce trades at agreed prices
• In practice, there is no guarantee that this happens: many countries have poorly developed court systems and corrupt governments who do these jobs poorly, widely studied in political economy, creates an impediment to having an effective market economy

Question 13

The Distribution of Property Rights in a Market Economy

Answer 13

• Behind a market economy is a distribution of property rights
• No slavery assumption: people own their own labour power
• But this does not mean that they own the output that they produce
• We will consider a distribution of property rights to output which could be different to who produces what
• Changes in the distribution of property rights to output corresponds it what economists call lump-sum redistribution
• Taking output from one person and giving to another
• Reallocation of property rights

Question 14

Edgeworth Box

Answer 14

• Two individuals A and B
• Two goods: books (good 1) and food (good 2)
• Individual A can produce food and individual B writes books
• But the distribution of property rights determines who owns the books and food
• Suppose there is a fixed labour supply (zero disutility of labour)
• If both like books and food then it would make sense for them to trade
• There is a market for books and market for food
• Let p1(p2) be the price of books (food)
• Income depends on poverty rights which are equal to 1

Question 15

EB, Two parts:

Answer 15

• Goods availability for each individual
• Reflecting the amount of food and books that each individual owns
• Distribution of income
• At an equilibrium point prices allocate consumption so that each consumer gets what they want given equilibrium prices
• We will now build up a picture of general equilibrium (supply = demand)

Question 16

General Equilibrium point:

Answer 16

• General equilibrium is the point where both consumers get what they want given the prices
• Demand comes from tangency between indifference curve and budget line
• Both simultaneously achieving tangency where supply = demand

Question 17

Wages and Prices in General Equilibrium

Answer 17

• Reflect supply and demand in different markets
• All else equal:
• Benefit as consumers when fewer people demand the things that they like
• Benefit as producers if?
• Wages and prices distribute who gets what
• Property rights decide who owns what

Question 18

The Utility Possibility Frontier (UPF)

Answer 18

• Now ask what level of utility does each consumer obtain in a general equilibrium
• This is a crucial idea when it comes to thinking about efficiency and equity
• We allow for the possibility of redistribution between the two individuals
• Imagine a redistribution of property rights to output
• Relative prices will change to reflect the scarcity of the goods to achieve a general equilibrium outcome
• And each will be associated with a particular level of utility for each individual
• Endowment: who starts in an advantageous position, what are the wages and prices?
• Increase utility of A, decrease utility of B
• All different levels of A and B depend on resource endowment
• A point on the frontier corresponds to a general equilibrium point

Question 19

Lump-sum redistribution

Answer 19

• What happens if there is a change of property rights so that some of the food produced by A is transferred to B before trade begins
• After the transfer of food, we assume that they trade in markets to achieve a general equilibrium
• Prices may now have to change to clear markets depending on the preferences of the two individuals
• Individual B who now has a greater purchasing power may like one good more than the other
• This results in a new point on the UPF
• So the UPF maps out what happens in a market system with different levels of purchasing power
• Important and useful when we think about redistribution
• Inequality cause by starting point
• Market only exacerbates inequalities

Question 20

Welfare Economics

Answer 20

• Branch of economics that engages with normative analysis and moral philosophy
• This underpins how we think about the role of government
• A cornerstone of the approach is a framework for thinking about inequality
• We will try to think systematically about welfare judgements based on the distribution of income

Question 21

The First Fundamental Welfare Theorem

Answer 21

• Consider an equilibrium price vector p* (p1,…,pN)
• Key concept: Pareto efficient
• An allocation of goods and services (x,L) is Pareto efficient if we cannot make someone better off without making someone else worse off
• The first fundamental theorem of welfare economics says: every general competitive equilibrium is Pareto efficient (supply = demand)
• But, nobody is directly coordinating economic activity and people are acting independently and selfishly
• The invisible hand: a term used by Adam Smith to describe the unintended social benefits of an individual’s self-interested actions
• Why is it true?
• We know that everyone is maximising given the prices that they face
• Then if there is bundle of consumption and labour supply that a consumer prefers, it must be unaffordable
• Equally if there is a consumption bundle with U(x,L) > U(f,g) then is must be infeasible
• Market is Pareto efficient by construction, in market, everyone is maximising if someone could be doing better they would be
• Not every point in an Edgeworth box is Pareto efficient: just need to reallocate
• Trade until everyone is better off

Question 22

Contract Curve

Answer 22

• The contract curve is the set of points representing final allocations of two goods between two people that could occur as a result of mutually beneficial trading between those people given their initial allocations of the goods
• Directions of Pareto change
• Any point on contract curve is Pareto efficient where trade has been exhausted
• Every transaction creates a Pareto improvement, assumes you are a ‘positively’ free consumer
• At the market clearing price, everyone gets what they want at the equilibrium point
• But, a society can be Pareto optimal and still perfectly disgusting
• Edgeworth box can’t say whether Ua or Ub is better or worse off, we would have to assume utilities matter comparatively??

Question 23

What determines the distribution of income in general equilibrium?

Answer 23

1) Distribution of property rights
2) Distribution of talent
3) Distribution of effort
4) Market prices

• With equal ownership, property rights is the same for everyone and every good, there will be equality of incomes
• If individuals own what they produce, then an individual skilled in the production of k gets income independent of property rights
• Therefore inequality must depend on
Level of skill - a
Scarcity - X
Extent of effort - g

Question 24

Distribution of utility

Answer 24

• Individuals may have different tastes for income and goods
• Some people may dislike work/value consumption more
• This will be important when we think about a just distribution of income looks like

Question 25

Partial Equilbrium

Answer 25

• An examination of equilibrium and changes in equilibrium in one market in isolation
• In partial-equilibrium analysis we hold the price and quantities of other goods fixed
• We ignore the possibility that events in this market affect other markets’ equilibrium prices and quantities
• At times partial equilibrium might understate the effect

Question 26

The market system does not guarantee equity

Answer 26

• Using a general equilibrium model
• All individuals in the economy have an occupation which is associated with the production of a specific good j
• Then aij is i’s capacity in producing j
• Then income is m = aijpjLi where we can think of wi = aijpi as like i’s wage
• It is obvious that there is nothing which equalises incomes when prices are determined by supply and demand

Question 27

Measuring Income Inequality

Answer 27

• Most statistical studies look at income inequality, based on samples of individuals in a population
• Complicated by the fact that most people live in multi-person households
• So look at household income, but households are different (e.g. have scale economies)
• So normally look at equivalised income (adjusting for household composition)
• OECD modified equivalence scale in use since the 1990s: assigns a value of 1 to the household head, 0.5 to each additional adult member and 0.3 to each child
• Income inequality is skewed to the right

Question 28

Income Distribution

Answer 28

• It is not a normal distribution
• It is skewed to the right
• Means median income is less than mean income
• Inequality is mostly explained by distribution of wages
Distribution of wages
• More skewed to the right than distribution of income

Question 29

The Lorenz Curve

Answer 29

• The Lorenz curve is the graphical / visual representation of income or wealth distribution developed by American economist Max Lorenz in 1905.
• It is drawn as a cumulative income curve.
• A population is divided into quintiles: The richest quintile is the 20% of households with the highest disposable income. Similarly, the poorest quintile is the 20% of households with the lowest disposable income.
• The Lorenz Curve can then be contrasted with the line of perfect equality to show the scale of income and wealth inequality in a country.

Question 30

The Gini Coefficient

Answer 30

• The Gini coefficient is a commonly-used measure of income inequality that condenses the entire income distribution for a country into a single number between 0 and 1: the higher the number, the greater the degree of income inequality.
• A value of 0 corresponds to the absence of inequality, so that, having adjusted for household size and composition, all individuals have the same household income.
• In contrast, a value of 1 corresponds to inequality in its most extreme form, with a single individual having all the income in the economy

Question 31

A Richer Model of Income Distribution

Answer 31

• Income distribution is a by-product of the market system
• Reflects behaviour and rewards in many markets
• (And, as we shall see in later lectures also depends on government behaviour)
• But labour markets are not the only source of income differences
• In a richer model of production firms also use capital, natural resources and technology based on an underlying production function
• Ownership of these also affects income distribution

Question 32

The importance of thinking in general equilibrium terms

Answer 32

• Rewards are largely determined in market equilibrium
• They differ across time and space due to distribution of property rights and relative scarcity of different of different skills
• These can be different in different locations and can change over time
• Also international trade can affect prices in an open economy
• Rising inequality is frequently attributed to globalisation which affects prices: lowering prices of some goods, lowering wages of those who produce them

Question 33

Gini coefficient across countries

Answer 33

• International comparisons are fraught with difficulty
• Countries tend to use household surveys which use very different methods
• E.g. different time periods for assessing income
• Accounting for all sources of income
• Gifts and transfers to such as housing
• Capital gains/losses
• Capital/labour income is very hard to measure and even tax
• But global trade doesn’t necessary mean increasing inequality (Turkey, Greece)

Question 34

Some Big Debates

Answer 34

• There are two key views of the market system

1) Markets exploit gains from trade and give people freedom
- This can mean accepting the inequalities that arise in the market place
2) Markets are inherently unequal because they accentuate underlying inequalities in talent/social background/property rights
- Begs questions of what to do instead of market allocation
- So this require us to think about the role of government in theory and practice

Question 35

Is income the same as welfare?

Answer 35

• It has the advantage of being measurable
• Although in more complex situations measuring income is not straightforward
• But our core theories are about utility not income
• For choice purposes utility is ordinal
• But for studying social welfare, we need to be able to compare the utility of individuals
• No presumption of comparability as utility is ordinal

Question 36

Can there be a science of welfare?

Answer 36

• Suppose there are policies which benefit some and make others worse off
• So they are not rank-able on the Pareto criterion
• We would like to know if the policy is worthwhile
• Could be small scale policy (traffic flow) or grand big debates (Brexit)
• Welfare economics analyses these issues

Question 37

The Nature of Utility for Welfare Purposes

Answer 37

• Ordinal utility means (behaviour of single individuals) if U(x) is a utility function then so if F(U(x)) where F(.) is an increasing function (have the same behaviour and indifference curves look exactly the same)
• We can only say something is better or worse but not by how much
• Cardinal utility means if U(x) is a utility function then so is a + bU(x) where a and b > 0 are constants (linear structure)
• Allows us to think about how much better or worse something is once we have fixed a bad b
• Ordinality is used in standard consumer theory to look at demand
• Cardinality is used in choice under uncertainty as the shape of the utility function is important
• Any claim about diminishing marginal utility requires cardinal utility
• E.g. Temperature -> ordinal: is it hotter or colder? (Subjective)
• Cardinal: Hotter or colder by how much? (Objective)
• Marking interpersonal comparisons is quite separate from ordinality or cardinality
• Take two individuals A and B and their consumption then can we say whether U(A) < U(B)
• This is required if we are going to make welfare comparisons based on utility
• Note that utility could be ordinally or cardinally comparable
• Ordinal comparability = we can say whether someone is better or worse off than some one else but not by how much (this is all that is needed for maximin
  
  • Cardinal comparability = we can measure how much better off or worse off someone than someone else (required for most “welfare” approaches)

Question 38

Income or Utility?

Answer 38

• Income is both cardinal and comparable across individuals.
• Cardinal since we can change the units of income (e.g. pounds, pence, euros etc)
• But the ranking will be the same
• Comparable since money is a common metric
• If all income is spent then m = sum of (px), income is directly related to consumption
• Higher income will therefore mean mor utility but we would not be able to say by how much without taking a stronger stance on interpersonal comparisons

Question 39

Utility and Income

Answer 39

• If nu > 0, there is diminishing marginal utility of income
• Marginal utility is a decreasing function of m
• This the assumption that most people believe in
• As income goes up, get less utility

Question 40

Can Utility be Measured?

Answer 40

• Happiness levels subject to short-term fluctuations
• Life-time satisfaction seems more reliable
• One popular approach is to use surveys of life satisfaction
• Respondents are asked to rate their life satisfaction on a scale of 1-10, policy makers take this data seriously
• There is no extensive evidence on utility measured this way and what it depends upon
• And income is a consistent factor driving life-satisfaction
• Although a lot of other things matter, e.g. U in shape

Question 41

Relationship between life satisfaction and income

Answer 41

• Estimate nu = 0.6

- Roughly linear in logs i.e. happiness increase with income but at a decreasing rate

Question 42

Prices and Utility

Answer 42

• Basic theory emphasises that prices also affect utility
• Utility naturally reflects that “real income” matters
• Prices affect “purchasing power”
• An increase in a relative price lowers “real” income
• Real income is measured by a deflating price index
• This is important in comparing incomes across countries and within-countries over time
• E.g. poor services have different relative prices (such as cheaper services)
• E.g. prices of goods changes over time (such as falling prices of computing power)
• Nominal price may not reflect fall but real price may (adjustment needed)

Question 43

Are you inequality adverse?

Answer 43

• The Pigou-Dalton Principle of Transfers
• Suppose that there is a fixed sum of utility
• And you transfer utility from a richer to a poor individual
• Then do you think that social welfare should be higher (make transfer but kept sum the same, could be done with income)
• Then you are inequality averse

Question 44

Relating Social Welfare Change to Policy Interventions

Answer 44

• This is one of the main motivations for having a metric like social welfare
• Cost Benefit analysis
  Pareto principle?
• Not much help as the training in investment is not likely to make either worker worse off
• So it would be indifferent
  Biggest increase in income?
• Would give us a determinate answer
• But would not naturally allow any focus on the worse off
  Biggest increase in social welfare
• Now we introduce a welfare weight which could be higher for the poor
• And the policy that raises income need not be the one that increases social welfare most
• Could appeal to diminishing marginal utility of income or an SWF with a preference for equality

Question 45

Concerns about utility?

Answer 45

• We have made life simple by assuming that people have the same preferences
• But suppose that one individual gets less enjoyment from a given utility than another
• U(A) = lambda U(B) with lambda < 1
• Then the marginal utility of income is also lower for A
• So A will get less resources than B
• Amartya Sen has argued that this is not right
• Most intuitions would say that it could depend on the reason that lambda < 1
• He formulated the weak equity axiom (example of able and infirm)
• If person A is worse off than person B whenever A and B have the same income level, then no less income should be given to A than to B in the optimal solution of the pure distribution problem (capabilities framework)
• Embodies a value judgement which conflicts welfarism

Question 46

What about rights and freedom?

Answer 46

• “Underlying most arguments against the free market is a lack of belief in freedom itself” Friedman
• One argument that is often made in favour of the market is that outcomes are the product of free choice
• Individuals are better off than they would be without trading in the market
• Hence the market is the product of choice and should be value for this
• This is a very different philosophical framework
• It is a process-based rather than outcome-based evaluation
• But often complained that the initial allocation of property rights and talents is a product of history
• See Wilt Chamberlain Argument (Justice in Transfer)

Question 47

Deprivation versus inequality

Answer 47

• Some people believe that there should be a focus on the well-being of the lowest income (Rawlsian approach)
• Motivates a focus on poverty
• But this could be absolute or relative poverty
• Absolute poverty: those with income below a cutoff level of income
• Relative poverty: those below half median income
• Poverty can also be about consumption of specific goods (e.g. energy, health care or food.)

Question 48

The Problem of Social Order

Answer 48

• How to establish a social order is the heart of many issues in social science
• People can take actions which affect the well-being of others:
Agreeing to trade with them
Acts of aggression and violence
Helping others out
• The market system as we have studied so far is a model of social order
• Individuals interact solely through markets with interactions mediated through market prices
• Bringing in public goods and externalities fundamentally alters the challenges involved in establishing a social order
• Raise fundamental issues around how a society is organised
• This is because interdependence in not well-managed via markets
• There are two main traditions:
1) Social contract:
• Social order is established through finding ways of creating organisational structures that facilitate cooperation
• Although this cannot be handled in markets, the metaphor of a contract between individuals still applies
• The problem is how to enable contract-like solutions to emerge
2) Hobbesian tradition (Force majeur)
• Social order needs stronger coercive authority
• Mainly focuses on government rather than other solutions to living interdependently
• We will explore both solutions below when we have introduced idea of public goods and externalities

Question 49

Types of Goods

Answer 49

• The goods that we have studied so far have two key properties
• Depletability - more for one means less for another
• Excludability - individuals benefit from consumption exclusively
• But many goods are not like this
• Note that this applies equally well to bads as well as goods
• Indeed there are examples where a good to one person is a bad to another (e.g. playing loud music)

Question 50

Living Interdependently

Answer 50

• Non-excludability is critical when people live together in an interdependent way
• We may not be able to choose the consumption of goods that enter our utility functions (e.g. local, when litter left on ground or global, in the case of carbon emissions
• We will explore the implications of this for a world where individuals make independent decisions
• As would be the case in a pure market system

Question 51

The Tragedy of the Commons

Answer 51

• Term coined by Russell Hardin (1940-2017)
• A situation in a shared-resource system where individual users acting independently according to their own self-interest behave contrary to the common good of all users by depleting or spoiling that resource through their collective action
• Many things that we consume are common resources
• Such as the planet’s environment or whether we live in safe neighbourhoods
• We take actions which affect how much common resource there if for others to share
• So we need to find ways of living together more effectively

Question 52

A Pure Public Good

Answer 52

• Suppose that there are N goods as in our corse model above but good 1 is special
• It is a pure public good
• With with individuals enjoying equally their own consumption and that of all members of society
• G is a pure public good and is a function of consumption for everybody
• We will focus on the case where G is as public good, i.e. U is increasing in G
• But then we could write many different models which capture a rich array of situations

Question 53

Example: The Need for Collective Security

Answer 53

• Two people face a common enemy
• Since they live in the same location, it is the joint effort to produced security which matters
• I.e. security is a public good
• Thus G is a measure of secure they are against enemy attack
• The parameter Y represents how much security matters which could depend on the strength of an external threat
• Providing security is costly
• It requires each individual to give up another activity which they enjoy
• Will they protect themselves or live in an insecure environment?

Question 54

Nash Equilibrium

Answer 54

• Understanding consumption choices now requires us to use the logic of game theory
• Interest between individual A and B is a game
• We will look for a Nash Equilibrium
• Each agent takes the choice made by the other agent as given and then chooses their best outcome
• At equilibrium both are doing the best they can given what the other is doing
• We explore this by constructing the payoff matrix associated with all combinations of actions by both individuals

Question 55

Public Goods Provision as a Prisoners’ Dilemma

Implications

Answer 55

• This is called the “free-rider problem” since each individual gets a benefit from the contribution of the other (non-cooperative behaviour) but pays none of the cost when the other person provides the good
• Don’t get enough utility directly from contribution
• The outcome achieved is Pareto inefficient
• A and B get 2 each
• Whereas with x1B = x1A = 1, they would each get 2.6
• So both could be strictly better off if they could make different decisions
• But this is not compatible with their incentives
• So if we think about the decision over x1i being made in a market, then the market fails

Question 56

The Idea of Market Failure

Answer 56

• This simple world of consumption of a public good illustrates the idea of a market failure: A market failure is where it is possible to make everyone better off with some other choices
• The first fundamental welfare theorem does not hold with a public good
• Why?
• When there is interdependent consumption, individuals cannot optimise over everything in their utility function
• The market does not allow me to choose my neighbour’s consumption, only mine
• And there is no reason to think that my neighbour will do what I would like
• Especially when they have to pay the cost of doing something which I like

Question 57

An Application

Answer 57

• Although the model is very stylised, there are many applications
• For example, the problem of climate change (Benefits global, actions down to individual countries)
• Think of i as a country rather than an individual
  
  • And x1i is restricted between {0,1} is consumption of low carbon alternatives which benefit the other country
• Then without some kind of compact, there would be no incentive to cut carbon emissions
• So left to their own devices carbon emissions will be too high
• The class will give an opportunity to discuss some other examples

Question 58

Solutions to the free-rider problem

Answer 58

• “Voluntary” cooperation - evolve ways of negotiating and/or punishing free-riders
• Government intervention - have government take over provision and/or financing of the good

Question 59

“Voluntary” Cooperation

Answer 59

• Small scale societies often use negotiation to solve free rider problems, good examples are households and families
• But the requirements are more demanding in the prisoners’ dilemmas than assurance games
• Societies find ways of punishing defectors or those who do not make the best choices
• This could be explicit or indirect punishments
• The use of punishments is one reason that voluntary is in quote marks
• Many organisations try to organise cooperative means of solving free-rider problems

Question 60

Is self-interest the right assumption?

Answer 60

• This is the classic economic assumption but perhaps less warranted when there is interdependence
• But there are many ways of relaxing self interest:
Altruism - individuals care directly about each other’s payoffs
Reciprocal preference - individuals care conditionally about the actions of others (willing to pay price to punish someone who has done something wrong)
• It is arguable that there is a biological basis in such preferences (e.g. public good problem in getting something to eat for birds)

Question 61

Evidence from Lab Experiments

Answer 61

• There is now a lot of experimental evidence in controlled environment which close fit the kind of framework that we have studied here
• A group of M individuals are given “tokens” which they can use to contribute towards a public good
• So self-interest says that [contribution by k] = 0 for all k ]
• Pareto efficient outcome is [contribution by k] = tokens for all k
• Therese experiments have been running thousands of times
• Typical contributions are between the self-interested outcome and the Pareto efficient outcome
• There is some evidence that this varies across cultures/sub-populations
• Contributions tend to decline when the experiment is run multiple times
• Some kinds of pre-experiment communication makes a difference
• The question is why this happens

Question 62

Reciprocal actions and free-riding

Answer 62

• Now suppose that individuals punish free riders
• A imposes a penalty of size pi if x1A = 1 and x1B = 0
• B imposes a penalty of size pi if x1A = 0 and x1B = 1
• This is known as negative reciprocity
• Can this solve the free-rider problem?
• There are two key assumptions
• It is possible to detect free-riding
• Punishments are possible
• Although it goes beyond what we are studying in this lecture, we can think of pi as future exclusion
• In many settings punishment can also take the form of withdrawing cooperation in future
• But studying this requires a model where cooperation takes place over time
• This again is realistic in smaller scale forms of cooperation

Question 63

The Coase Theorem

Answer 63

• Named after Ronald Coase (1910-2013)
• Suppose that individual A owns the public good
• So it can be provided only if they say so
• But they can collect contributions from B
• A offers a contract to B comprising a level of public goods depending on how much B contributes {G, x1B} which may implicitly include a promise of their own contribution: x1A
• What is going on?
• The market system solves the problem once ownership rights are established
• Although requires a legal framework to enforce the contract
• The legal threat of zero public goods provision can induce private contributions
• The Coase Theorem is generally used to argue that we do not necessarily need government if property rights can be defined
• But probably best thought of as a different way of thinking about why public goods are different
• However, there are example where the Coase theorem could apply? (“Lighthouse in Economics”)

Question 64

The Role of Government

Answer 64

• It is arguable that a lot fo what government does is to interfere in markets for public goods
• What government brings to the table is the possibility of coercion
• Consumption could be regulated when it has benefits to others (e.g. compulsory vaccination)
• Or the government could take over decisions about the provision of public goods (tax = coercion) -> defence
• In the later cases it can use its coercive taxing power to fund public good provision
• Whether government uses its coercive power for public benefit is an issue that we will explore later
• For the moment, we will explore the role of government under the assumption that government cares abut social welfare

Question 65

Directly Regulated Provision

Answer 65

• The government could mandate in the prisoner’s dilemma example that x1A = x1B = 1
• But that mandate assumes that government can both observe x1i and has sufficient coercive power
• We will suppose that the government has the ability to issue a fine for setting x1i = 0
• Trying to regulate private provision may not be easy
• For example it may be very difficult to monitor x1i
• So now suppose that the government can levy a tax on everyone and use this to fund G
• The government collects taxes from income at rate t for everybody and chooses how much to spend on G
• Let us begin with the linear example and ask when the government will optimally spend
• Suppose that there is a maximum tax rate T above which people will evade their taxes
• I.e. t < T
• Suppose that the government is Utilitarian
  • I.e. it looks at the sum of the benefits of the public good across the population
  • Indeed if this holds, the government will set t = T, maximal provision of the public good
  • This is called the Samuelson rule for provision of a public good
  • (After Paul Samuelson (1915-2009)
  • This argument is highly intuitive
  • The value of reduced private consumption (good 2) has to be compared to the increase in consumption of the public good
  • The public good has to be paid for with an income tax, this means reducing each person’s income, and hence consumption of good 2, proportionately
  • This is illustrative only and other tax systems could be studied
  • The public good benefits everyone
  • So valuation is M x y not y as in the case of a private good

Question 66

Bringing in Distribution

Answer 66

• This is already implicit in what we have looked at as there is diminishing marginal utility of private consumption
• But we can do this more explicitly using a social welfare function
• It is also interesting to look at government intervention to provide public goods when incomes differ
• So with an income tax not everyone pays the same amount (m differs)
• We could also have differences in benefits from the public good (lambda/y differs)
• We can also relate provision of a public good to the Utility Possibility Frontier
• Develop a two person example
• Incomes are such that ma > mb so that with an income tax, A pays more than B
• Assume still that y is the same for both individuals
• In the absence of government, we know that there will be no public goods provided
• Allow for a social welfare function which values utility differently

Question 67

Key Observations

Answer 67

• There is no guarantee that with government intervention, every citizen is better off with intervention than without
• With a preference for equality, government intervention can increase social welfare even if one agent loses out
• This is because the social welfare function weights the gains of the worst off individual more than the better off individual
• But this may mean trading off losses against benefits
• If some citizens are better off and some are worse off, then politics is going to matter

Question 68

Take Home

Answer 68

• Once consumption based patterns become intertwined, then independent market-based decisions are generally not Pareto efficient
• Thus the first fundamental welfare theorem fails
• Markets are not efficient (most widely used definition of market failure)
• Responding to this varies by context
• Solutions with small groups can use local information and social punishments
• For larger scale public goods, the state generally gets involved
• But then there is no guarantee that there is a Pareto improvement over the no-state outcome
• So the case for government intervention has to be considered case by case and politics matter

Question 69

Behavioural Economics

Answer 69

• “A method of economic analysis that applies psychological insights into human behaviour to explain economic decisions-making”
• Or economics with “real humans”
• Provides a way of bringing economics and psychology closer together
• Has already resulted in two Nobel prizes
• Will explore some key ideas and their implications
• Heavily motivated by lab experiments
• But there are many different phenomena which travel under the heading of “behavioural economics”
• We will explore only a few of the ideas in what is now a huge area

Question 70

Hot and Cold States

Answer 70

• Preferences vary according to psychological (hot/cold) states
• Hot and cold refer to emotional not physical states
• Effect of emotional states: getting excited, feeling down
• Being distracted by payoff irreverent cues, i.e. you associate the colour of an object with something nice
• Acting on instinct: using intuitions rather than thinking it through
• The hot state is like ordering food in a restaurant when you like the look of something unhealthy
• The cold state is when you are eating and it gradually dawns on you that I should be eating healthy and hence regret your order
• So you have made the “wrong choice”
• This is because my “decision utility” differs from my “experience utility”
• And your well-being is based on your “experience utility”
• You would like to make a decision in the cold state but you cannot always rely on that
• The world is arguably full of examples where this idea applies
• You can think of having two selves:
• The “hot self” who likes good 2 and the “cold self” who likes good 1

Question 71

Welfare

Answer 71

• If decision making is in the hot state and consumption in the cold state, the consumer is in trouble
• Cold state utility is the true “experience utility” and
• So the consumer makes a mistake by deciding in the hot state
• And x2 = 0 is the Welfare optimal choice
• Bringing it about creates a Pareto improvement
• This gives rise to a new potential set of arguments for government intervention
• Although this requires that someone knows better than the consumer what is good for her

Question 72

Implications for Micro-Economics

Answer 72

• Free choices made in a “market” need not be “optimal”
• We may end up regretting our purchases (sub-optimal choices)
• Worse still, markets can take advantage of the way you make decisions
• Creating “tempting” product is which are not good for you but you buy anyway
• Advertisers understand how to frame-decisions to make you like products (creating imagery to enhance your decision utility)
• And some businesses use high-pressure sales tactics to get you to buy something before you have thought it through
• Are there market solutions?
• Finding ways of avoiding decisions in the wrong context
• People often make it hard to access their savings
• And there are some products in the market that help them to that
• There is evidence that people understand this and prefer “commitment savings products”
• Are there government solutions?
• Banning/taxing some products
• Taxes high sugar drinks
• “Cooling-off” clauses which allow you to change your decision
• UK government gives a right to individuals to cancel certain transactions such as motor insurance and consumer loans without penalty within 14 days

Question 73

The Value of Commitment

Answer 73

• In the above example, your cold self would like a way to stop you choosing x2
• But this means finding a way of preventing this decisions when it is relevant
• E.g. maybe you choose a restaurant that only serves healthy food, or leave your phone in a different room

Question 74

The Example of savings

Answer 74

Do you consume now or later?
• Being young and enjoying consumption is like being in the “hot state”
• People who get to old age and wish that they had saved more when they were young

Question 75

A Commitment Savings Product

Answer 75

• It restricted access to their savings, although each individual could define a goal date or amount
• For the commitment savings group, average savings balance increased by 42 percent after six months and by 82 percent after one year compared to the others

Question 76

Internalities and Government Intervention

Answer 76

• We can think of the problem as an “internality” to parallel the idea of an externality
• But it is your own “future self” that benefits not someone else
• The behaviour of a consumer in the hot state is lowering the utility of her “future self” in the cold state
• One possibility would be to tax good x2 or even ban it?
• Although banning is an extreme solution
• Probably only persuasive where the good is really bad for you
• This could be done on the basis that the state knows best and acts in the interests of its citizens
• Requires us to believe that government is benevolent
• Also it can create distortions for “responsible consumers”
• (Playing a game against your future self)
• Unlike B, consumer A does not know what is bets for her and this consuming good x2
• Then there is a trade-off
• Taxing or banning hits a consumer whose experience utility is higher with good 2
• We now need a social welfare function to help us decide what the best course of action is
• No Pareto improvement is possible
• Consumption decisions distorted
• Harming rational consumer at expense of irrational consumer
• So the ban is always justified if the aim is to protect only the worst off
• This might seem like a paternalistic stance but it is actually a distributional judgement
• The form of the social welfare function does matter
• Who are the vulnerable consumers?
• Are behaviour biases linked to age, poverty or low education?
• Then intervention could be justified by helping these more deprived groups (they consume at higher prices)
• Poverty may also create more stress which impedes cognitive ability
• So the poor may be more inclined to make faulty decisions with damaging consequences
• Undermining case of the market if poorer people buying more expensive goods (market failure)
• What about targeted interventions?
• Is there a way of influencing B without distorting A?

Question 77

Nudges:

Answer 77

• Create a default option of x2 = 0
• Allow B to opt out to choose x2 > 0
• Works if the opt out helps B to make the choice in the cold state
Example:
• Opt out retirement savings, individuals have to make a conscious choice to exit the scheme

Question 78

Behavioural Economics Summary

Answer 78

• Explored a framework where consumers do not use their “true” utility function when making decisions
• This has welfare consequences
• In principle, getting them to make the correct choices can lead to Pareto improvements
• But not always if some consumers are not biased
• There may be “marketed” solutions that help people improve their decisions (e.g. commitment products)
• There are therefore trade-offs involved in government intervention:
• Is it worthwhile distorting non-biased consumers to help biased consumers
• We need a social welfare function to help us decided whether to intervene and how

Question 79

Decision Making Under Uncertainty

Answer 79

• Introduce idea of expected utility
• Also: over-optimism, reference-dependent choice
• Many decisions in economics are made without full knowledge
• We can extend our models of behaviour to incorporate this
• It will allow us to think about aversion to risk and what judgements are needed to make choices
• Although choice under uncertainty is generally used quite loosely, it is useful to make a distinction between:
• Risk: facing an uncertain but quantifiable future
• Uncertainty: facing an uncertain and unquantifiable future
• (E.g. prior to financial crisis, bankruptcy of large bank assumed p = 0, unquantifiable (ignored) risk)
• Idea of unknown unknowns

Question 80

Quantifying A Risky Future

Answer 80

• This is straightforward in principle
• Step 1: Think of all the events that might happen: x1, …, xk
• Step 2: Attach probabilities to each event: p1, …, pk
• Although note that people may have different information about this and may also be able to take actions which affect the likelihood of different events occurring
• Specifying events and probabilities is what an individual believes about the future
• People can have different views about what is going to happen
• For example, scientists may have a different assessment of climate change that most people “in the street”
• Sometime probability is based on an objective process
• E.g. a fair coin has 0.5/0.5 probability but many things are not like this
• There is a lot of scope for psychology
• Cardinality means that the shape of U(xk) matters
• So there can be diminishing marginal utility (dmu)
• Risk aversion= dmu

Question 81

Risk Aversion?

Answer 81

• To study this, we focus on the case where events are different levels of income
• So individuals are exposed to risk in the income profile that they face
• This could be by choice or because they face income shocks (e.g. farmer who faces wether shocks)
• The framework could in principle by applied to any kind of risky events (e.g. shocks to health)
• We will now let U(m) be the utility from income
• Will ask whether people like or dislike income risk
• Sine utility is cardinal, it is meaningful to talk about the shape of U(m) e.g. concave, convex etc.
• We specialise the model to consider a binary choice problem

Question 82

The Demand for Insurance

Answer 82

• How much would an individual be willing to give up to achieve certainty?
• This is important in the insurance industry since people buy insurance so that they reduce the risk that they face
• So there is a close link between the demand for insurance and risk aversion
• We can see this by thinking about an insurance company selling a product to a risk-averse individual
• This makes sense if insurers can pool risks across over many individuals So behaves as if they are risk-neutral
• Insurance company pools everyone’s risk
• Both are related to the idea of concave utility
• And Rawls motivated the idea of being inequality averse through the idea of a choice behind a veil of ignorance
• It is risky since you do not know where you will end up
• Hence you choose your redistributive preference “as if” you are risk averse which makes you averse to inequality
• Rawls went to the extreme and proposed maximin which is extreme risk aversion

Question 83

Choosing Levels of Risk

Answer 83

• One way to think about increasing risk is to suppose that two distributions have the same mean but wider variance
• More generally a distortion is riskier than another if it is mean preserving spread??k
• One distribution adds “pure random noise”
• A risk averse person will always prefer the less risky distribution

Question 84

Back to Behavioural Economics

Answer 84

• The expected utility approach is conventional
• Fertile ground for behavioural economics, this lies in part because of the anomalies in decisions-making which seem to violate the core model
• The key assumption in expected utility theory is the idea that we can seahorse beliefs and preferences in a simple way
• p1, … pk: beliefs (probabilities of k possible events)
• U(mk): preferences (Experience utility)

Question 85

Anomalies

Answer 85

• People are not good at probabilistic reasoning
• E.g. framing, overconfidence, exaggeration of small risk and neglect of big risks
• Their preferences also do not seem to conform well to standard models of U(m)
• E.g. reference dependence, loss aversion

Question 86

An Example: Framing

Answer 86

• People respond to the way that risks are described
• Describe an operation to a surgeon and change only one piece of information:
• 80% chance patient will live
• 20% chance patient will die
• Framing can have real effects when governments make risky decisions

Question 87

Prospect Theory

Answer 87

• This was developed as an alternative way of thinking about choice under uncertainty and was developed by Danny Kahneman and Amos Tversky
• Superficially, it looks very similar to expected utility

Question 88

The Value Function

Answer 88

• Kahneman and Tversky postulate that the value function where the outcome xk is evaluated relative to a reference point r
• Let xk - MPK be income then:
• V(mk) = v(mk - r)
• Where greater weight is placed on losses than gains
• This is due to evidence of the endowment effect: in experimental evidence by Kahneman, Knetsch and Thaler (1990) found that people place a higher value of a good that they own than they do not own
• Symptom of loss aversion, people value things relative to some point that they define
• Don’t take away policies people are used to?

Question 89

Example of Value Function

Answer 89

• Consider Olympic competitors
• Someone who expected and won bronze could feel better off than someone who won a silver and expected gold
• I.e. v(0) > v (silver-gold)
• Or someone expected gold could be worse off than someone winning silver when expecting bronze
• I.e. v(0) < v (silver-bronze)

Question 90

New York Cab Drivers

Answer 90

• Research by Camerer et al (1997) found evidence that taxi drivers in New York City appear to supply labour according to reference point in the form of “target level” earnings
• Their labour supply function slopes down
• If their wage per hour goes up, they work less as they reach their target earnings with fewer hours
• Can be rationalised by a Value Function

Question 91

Where do reference points come from?

Answer 91

• Bringing in reference points raises lots of issues
• Reference points have to come from somewhere
• Is it experience of decision utility?
• Context specific?

Question 92

The Value Function

Answer 92

• This can cause people who are apparently risk neutral (in their experience utility) to behave as risk averse
• Value gains less than hate losses

Question 93

Gambles with Reference Dependence

Answer 93

• People will be risk averse with reference gamble around reference point?
• But risk neutral with respect to gambles above and below the reference point
• Reinforces that the reference point can matter
• If we can influence reference points, this can have a real effect of choice under uncertainty

Question 94

The Effect of Reference Points (the Homer Simpson Theorem)

Answer 94

• Two students are deciding whether to work a bit harder to get a first
• One student is expecting a first and the other a 2:1
• Who is more willing to do the hard work when there is a 50:50 chance of getting a first or a 2:1 (p = 0.5)
• So the student expecting a first is actually less likely to get one
• “The secret to happiness is having low expectations”
• If reference points are not part of experience utility and we care about that for welfare purposes, then they are a distortion on otherwise optimal decisions
• If there are other distortions, perhaps reference dependence can overcome them?

Question 95

Overconfidence

Answer 95

• “It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so”
• Believing things that are wrong (with great confidence)
• Overconfidence can have a material effect on decisions that people make
• Examples (titanic, Chernobyl etc.., sub-prime mortgage crisis)
• Basic idea is that people have beliefs that put too much weight on events that they believe that they “know” and their decisions reflect this
• Can be thought of as “re-weighting” probabilities - this may better capture what drives behaviour

Question 96

An Example (of overconfidence)

Answer 96

• A risk-neutral individual who cares about income has to decide whether to make an investment which costs 2
• The investment yields 4 with probability 1/4 and 1 with probability 3/4
• So the gamble is a bad idea based on an expected value calculation:
1/4 x 4 + 3/4 x 1 = 1.75 < 2
• If the individual is overconfident and thinks that the probability of 4 is 1/2, then they evaluate the gamble as
1/2 x 4 + 1/2 x 1 = 2.5 > 2
• This would imply that on average in populations with overconfident individuals, investment returns will be lower
• This kind of bias was often used to explained the financial crisis
• Investors were over-confident about the returns on the financial products that they were creating/marketing

Question 97

Policy Responses

Answer 97

• More gov interventions
• Is government immune to behavioural biases?
• Policy-makers are human too but hopefully have expert opinion
• Do we educate people on overconfidence or tax/ban some gambles
• Increasing respect for theory that market failure is rooted in psychology
• Following the financial crisis, a lot of measures were put in place to make the financial system “safer”
• Greater regulation of some financial products
• But there was an extra element
• Not only was there overconfidence, but they were gambling with other people’s money, creating a major moral hazard problem (losses to investors not individuals)

Question 98

The idea of Political Failure

Answer 98

• Main government functions are taxation, regulation and spending
• What is gov failure?
1) Distribution issues:
• Do the distributional consequences of government action conform to reasonable normative criteria?
• I.e. a social welfare function
2) Efficiency issues
• Does the government spend/tax/regulate efficiently
• Can there be Pareto inefficient

Question 99

Example: The Problem of Corruption

Answer 99

• Government officials may take bribes and/or favour their friends/family
• Leads to arbitrary forms of redistribution via the state: towards those who have “political connections” who can skew policies away from those with broad societal benefits
• Can increase the costs of delivering public projects - poor quality public infrastructure delivered at high cost
• Creates poor societal incentives, entrepreneurs become rent-seekers
• Rent-seekers - capture rents of corrupt process/state instead of private sector
• Improves welfare of someone at the expense of the welfare of someone else

Question 100

Aggregating Preferences

Answer 100

• People have different views and interests
• The process of aggregation determines whose views/interests are reflected in policy
• This depends on the structure of political institutions
• For example elections serve to aggregate preferences
• This topic operates in the spectre of Arrow’s theorem
• Can a consistent social welfare function be derived from a set of agreeable procedures, if so can we employ them and justify they use of that social welfare function?
• Result of this became known as Arrow’s impossibility theorem

Question 101

Basics

Answer 101

• Consider a set of possible policies g1, …gv with typical element gv
• Suppose that we have a collection of individuals i with preferences ui(gv)
• We want to find a way of ranking the policies based on individual and then we might want to use that rule to make policy decisions
• Then Arrow said that a reasonable aggregation process ought to satisfy four reasonable conditions
• Begin with three obvious ounces and then one that is slightly trickier to grasp and less intuitively appealing

Question 102

Three Easy Ones

Answer 102

• Unrestricted domain (U)
• The procedure should work for all policies and configurations of preferences (be able tor and all policies)
• Non-dictatorship (D)
• We cannot have a ranking that is simply one person’s ranking
• Pareto principle (P)
• If everyone prefers one policy to another, then the ranking should prefer that policy to another
• All of these seem uncontroversial

Question 103

Independent of Irrelevant Alternatives

Answer 103

• IIA: The ranking between two policies ga and gb should depend only on the individual preference ranking between a and b (must be a pairwise comparison)
• Otherwise think of a jury deciding which of three people to award a prize to
• It decides to rank them 1,2,3 then adds up the score (Borda count)
• This violates IIA as if one judge changes their ranking between two candidates, it can change ranking between two other candidates
• The order is Candidate 1(5) > Candidate 2(6) > Candidate 3(7)
• Now suppose that Judge 2 switches their ranking between 1 and 3
• This violates IIA since changing in ranking between 1 and 3 by one judge changes the ranking between 1 and 2
• Rules out Borda count

Question 104

Arrow’s Theorem

Answer 104

• Impossibility theorem: A proof that it is impossible to devise a constitution or voting system, complying with certain reasonable conditions that can guarantee to produce a consistent set of preferences for a group from the preferences of the individuals making the group
• There is no consistent rule for the V > 2 which satisfies U, D, P and IIA

Question 105

Application to the Provision of Public Good: A Median Voter Result

Answer 105

• The public good can be provided by taxation at a per capita cost of c
• Each has the same income of m
• How much public good does each group want? (optimise)

Question 106

Downsian Politics

Answer 106

• Applied logic of markets to politics
• Two competing parties want to win an election and so offer alternatives to voters
• Voters choose which one to vote for
• Party strategies must form a Nash equilibrium
• Label the two parties 1 and 2 competing to win the election
• Preference rankings over alternatives
• Group A likes policy GA the most, where GA is the lowest level of public good
• Group C like policy GC the most, where GC is the highest level of public good
• Group B’s preference will prevail
• Suppose that a party proposes GB
• GB will always command 2/3rds of the vote against any other alternative
• So GB is a Condorcet winner

Question 107

A Median Voter Result

Answer 107

• So the outcome is in the middle, the medial value of GA, GB, GC
• And both parties converge to offering the same thing
• This is because preferences are “single-peaked”
• I.e. A and C prefer B to any of the extremes
• This is domain restriction in the sense of Arrow’s theorem
• This is often held up as a good “prediction” for how politics works (but it is a bit simplistic)
• However, it captures some of the basic logic about how moderating will be a viable electoral strategy

Question 108

Is the Median Voter Outcome a Welfare Optimum?

Answer 108

• So welfare wants the outcome for the mean public goods preference
• Politics delivers what the median voter, B, wants (they need not be the same)
• The proposition is true in general for any welfare function unless be chance, the welfare function is perfectly aligned with the median voter
• But this is a purely distributional difference, it affects who gets what
• This reinforces the idea politics will not generally maximise a social welfare function
• And we could regard this as a “government” failure if there is a big difference between what a welfare function says what we want and what politics delivers

Question 109

Quiz custodiet ipsos custodes?

Answer 109

• A fundamental problem of politics is making sure that people exercise power responsibly abuse of the state for private gain is a real issue in many countries
• Previously assumed that all public spending is spent on public goods
• But government can “selectively reward” groups that are close to the ruling elite, creating potentially serious problems of government failure
• Institutions provide the constraints on power that make government work in the public interest
• North Korea v South Korea as seen from out of space shows that institutions do matter!

Question 110

Motivation in Public Office

Theoretical Example

Answer 110

• A government has a source of tax revenue per capita of T
• This reflects the coercive power of the state
• It can spend recourse either on:
1) Transfer to the ruling elite: B
2) A public good which benefits all citizens: G
• The government’s budget constraint is: B + G = T
• It cannot spend more than it has in the form of tax revenues
• We will now suppose that citizens are identical and value public goods equally so we do not worry about preference aggregation across citizens

Question 111

Welfare Consequences

Answer 111

• The outcome is Pareto efficient with fixed T
• There is no ways of making both the citizens and ruler better off
• So if there is gov failure it must be based on a distributional criterion/social welfare function
• Some think that B is a social bad, i.e. as a matter of principle rulers should not use tax revenue to enrich themselves
• But could appeal to a social welfare function which dislikes inequality generated by the state
• Especially if political elites are on average richer than other citizens
• Distribution from poor to rich (regressive)

Question 112

Building Institutions

Answer 112

• Governments over history has evolved way of constraining rules in law
• Formation of Parliaments
• Creation of independent courts
• Free media
• Organised civil society
• Will focus on citizen representation through a parliament as a means of getting a better outcome for citizens

Question 113

Manga Carta

Answer 113

• Perhaps the most basic change in Constitutional history has been empowering citizens in tax raising
• The essence of the manga carta

Question 114

Key Ideas

Answer 114

• The citizens are represented in a Parliament which “bargains” with the ruling elite over how public resources are spent
• The rules can only raise additional taxes if parliament agrees
• So T is the tax revenue that can be raised without Parliamentary approval
• Could, for example, be government ownership of natural resources and other assets
• Citizens use Parliamentary Institutions to offer to pay more taxes in exchange for more public goods
• It strikes a social contract between citizens and the ruling elite
• This can result in a Pareto improvement

Question 115

When is a Pareto Improvement Possible?

Answer 115

• We need to see whether the utility of citizens is higher when they pay more tax as long as that tax is used to finance public goods
• They get G + R in public goods now and pay a bit more tax which reduces their income
• Is V(.R) increasing in R?
• Intuitively, the government is not spending enough on public goods
• Citizens are willing to pay more tax as long as this is spent on public goods
• So this additional taxation is voluntary
• If public goods are valuable enough, this will be the case
• To agree, the rule has to be no worse off, so a Parliamentary Institution facilitates this kind of bargaining arrangement

Question 116

Do people pay their taxes voluntarily

Answer 116

• Is it justifiable to chase on your taxes if you have a chance?
• 63% say no
• Those who justify cheating are; increasing in income, decreasing in age and eduction, tend to be men and have a lack of confidence in gov
• Just 200 voluntary payments of extra tax received in UK since 2000

Question 117

The Social Contract

Answer 117

• Citizens can offer to pay additional taxes R, so total taxation is now T + R
• In general the new levels of public spending are: (B’, G’) = T + R
• The ruler has to agree to this proposition as well as the citizens
• So if an agreement is reached, it must be a Pareto improvement over the status quo

Question 118

Implications

Answer 118

• So we have shown that there is scope for using institutions to improve the working of government
• Representative institutions can create a Pareto improvement
• Although that does not mean that reducing B is not also a legitimate goal on social welfare grounds
• Implicit in the way we have conceptualised this is that the Parliament “represents” the citizens
• So it tied to having representation in the form of elections
• And many societies also elect the “executive”, i.e. those who we are calling the rulers
• An alternative would be to rely on having those in public office be publicly spirited, but we need to be sure that we can find such people and select them

Question 119

Moral Hazard

Answer 119

• The problem of moral hazard is pervasive in economics
• In insurance contracts, insured people have an incentive to take more risks
• In employment relations, your employer or the shareholders bear the cost of poor performance
• In finance traders take risks on behalf of investors

Question 120

Asymmetric Information

Answer 120

• Generically, moral hazard and adverse selection are problems of asymmetric information
• One economic actor knows something another party does not know
• It is costly and possible impossible for everyone to know everything
• So we want to understand whether markets work well or not in such situations
• It can have implications for both efficiency and equity

Question 121

Market versus Government Solutions

Answer 121

• Sometimes markets can provide their own solutions
• For example, some products comes with guarantees so you can return the product if it is fault (the market provides its own reassurance)
• Incentive pay and financial penalties: some people are paid more if they are successful, reduces poor performance
• Bot there may be government solutions, governments establish trading standards, so firms are fined or customers are compensated when things “go wrong”

Question 122

Moral Hazard and Principal Agent Problems

Answer 122

• Two parties: 1) Principal who designs a contract and 2) Agent who is contracted with
• We are interested in the design of optimal contracts where the actions of the agent cannot be monitored directly, this is the essence of moral hazard
• For simplicity, we work with a risk neutral principal and agent i.e. who care about their expected income

Question 123

Framework

Answer 123

• Assume that the agent has no wealth so cannot be punished (financially) for failure, but can be rewarded for success
• To perform the task, the agent is therefore offered a “two part” contract: {b, w}
• w: a fixed wage just for showing up
• b: a bonus for being successful
• They also have an “outside option” with utility u if they work elsewhere

Question 124

Full Information Benchmark

Answer 124

• Suppose that the principal can specify e in a contract
• Then there is no need to use bonus pay, b
• So all remuneration is in the form of a fixed wage which has to be enough so that the agent participates in the contract

Question 125

Incentive Compatibility (Leonid Hurwicz)

Answer 125

• This refers to the fact that, even though, there is asymmetric information, contracts must anticipate this, so the contract is compatible with the incentives that the agent faces

Question 126

Illustrative example: A Publishing Contract

Answer 126

• The principal is a publisher and the agent is an author
• The payment w is an advance for writing a book
• But the book will only be a success if the author works hard
• Then b is the share of the royalties that the author gets if the book sells well
• And the publisher understands the relationship between the royalty share and the author’s effort
• The publisher also wants there author to write the book and/or not sign with a different publisher
• Note that w should be set as low as possible
• All pay should be in the form of “pay for performance”
• Would not be the case if the agent has “subsistence concerns”
• But makes the algebra simpler if we set w = 0

Question 127

The Optimal Contract

Answer 127

• There will be two cases:
• Case 1 where the outside option is not binding
• Case 2 where the outside option is binding
• The principal is willing to let the agent be better off than working elsewhere because there is an incentive benefit from doing so
• This is related to the idea of efficiency wages in macro
• Principle payoff is maximised
• Agent pay-off > outside option
• Pareto efficient, can’t increase one without making the other worse off
• The fear of losing your employees. means that you push up their compensation
• An argument that has often been used to justify high incentive pay in the financial sector
• If pi/2 < u, agent prefers outside option, therefore bonuses have to be pushed up
• Potential transfer of all profit to successful workers, generates more inequality as emphasises success and failure
  • So with moral hazard, there is a value from using performance-related contracts
  • They are dependent on observable success and failure
  • So only applicable where this is possible
  • Otherwise it is not possible to incentivise good performance
  • This matches with the extensive use of performance pay across the economy
  • This naturally generates inquests based on luck rather than effort when e<1

Question 128

Competition for Agents?

Answer 128

• Suppose that there are two identical principals competing for a single agent
• For all b < pi (profit?), the principal makes a positive profit
• So b will be pushed up until the principals earn zero profit
• At that point, e = pi/c
• So the inefficiency associated with moral hazard goes away with sufficient competition • But note that with more agents than principals, there is no competition and if the alternative is unemployment then e = pi/2c

Question 129

Welfare Cost of Moral Hazard?

Answer 129

• There are good reasons to expect economies with moral hazard to work at lower efficiency levels
• Expected output is lower
• But the outcome is Pareto efficient given the information problem (despite being inefficient in output terms)
• For this reason, this often referred to as “constrained Pareto efficiency”
• The principal and agent are “sharing” pi (profit)
• And some agents are successful (earn b) whilst others “fail”
• Such inequality is a reflection of the “optimal reward structures”
• It is a form of economically “jhsutificable” inequality to preserve incentives
• So society faces a potential equity-efficiency trade-off

Question 130

The Use of Bonus Pay

Answer 130

• Bonus pay is widely used but varies by sector
• We would expect bonus pay to vary by role and industry
• Bonus pay would vary by extent of moral hazard by cost of effort and measurability of output

Question 131

Does Bonus Pay Work?

Answer 131

• Three economists conducted a randomised control trial where incentives are randomly assigned to fruit pickers
• Allows them to compare the treated group with a control group
• Looked at fruit-picking performances
• Find that incentives are indeed effective but increase inequality
• Treatment group = incentives
• Control group = fixed wage
Results:
• Everyone better off by income inequality has increased
• Those in the bottom of the distribution have moved up
• Shifting middle to the tails

Question 132

Moral Hazard in Teams

Answer 132

• Frequently output it the result of many people working together
• Suppose that there are two team members A&B
• Success depends on both eA and eB
• Probability of success is E(eA, eB)
• Only teams can be rewarded
• So rewards have to be dependent on team outputs
• Rewards like public good
• Success is a public good
• Must overcome free-rider problem
• Peer pressure and private punishments could be important - you get to know your co-workers
• These may make incentives very effective

Question 133

Wealth Distribution

Answer 133

• We have supposed so far that the agent has no wealth
• But some interesting possibilities are opened up when this is not the case
• Agents can be pledged their wealth against failure
• Post a “bond” against poor performance
• In credit markets this means posting some collateral (if don’t pay money back, pay back as collateral)
• The principal can the use both carrots and sticks in solving inventive problems due to moral hazard
• However collateral doesn’t work if lender owes more than value of collateral (e.g. housing market crash)

Question 134

Intrinsic Motivation

Answer 134

• Moral compulsion
• Many psychologists and some economists argue that peopled do things “for the sake of it” and not just because they receive a monetary reward
• Worst still if you pay people, they mat actually resent it and do a worse job - known as motivation crowding out (e.g. if guest offers to pay for home cooked meal)
• If this is true, then the standard principal agent view of the model could need modifying
• Paying someone may demotivate them
• So the agent will work a certain amount “for free” but the willingness to do so may decline with earnings (likely to be highly contextual)

Question 135

Why Intrinsic Motivation?

Answer 135

• Offering you a reward makes you think that the task is more dangerous or difficult than you initially thought
• So personal benefit is a reflation of your view about this
• Being paid for something undermines your sense of self-determination
• Remind you that you are doing something for someone else rather than yourself
• You regard some tasks as in the non-monetary domain
• Completed out of a sense of loyalty, friendship or commitment to a good cause
• Being paid to do something cheapens the motive
• You want to signal something about yourself, e.g. conforming to a stereo-type that you identify with (being cool or nerdy)
• Some firms may invest in intrinsic motivation
• Reduce incentive pay from pi/2

Question 136

Summary of Intrinsic Motivation

Answer 136

• Blends ideas from psychology and economics to recognise that people do things for complex reasons
• Standard economic approaches that focus on incentive can backfire if the psychological summations about human motivation are wrong
• Incentives should probably be used more cautiously than in some economic settings
• Probability not fruit pickers but maybe for more vocational roles such as teachers/doctors etc.

Question 137

Summary of Moral Hazard

Answer 137

• Moral hazard has implications for the productivity of an economy
• It conforms with evidence that incentives do seem to work (most of the time)
• In general, it will leads people to put in less effort than desirable so solving moral hazard problems can increase productivity
• But does not automatically create a Pareto improvement

Question 138

Fundamentals (innovation)

Answer 138

• Understanding incentives to innovate is an important topic in micro-economics
• There are two main dimensions of innovation:
• New goods and services: product innovation
• Making existing things more cheaply: process innovation
• Successful innovation requires investment which is inherently risky
• Innovation takes place in firms and is also supported by government who subsidise private investment
• Many kinds of innovations are a public good
• Although there are laws which allow inventors to “protect” their ideas through patents

Question 139

Where does innovation come from?

Answer 139

• Invention and basic knowledge
• Development into useful products
• Commercialisation
• Improvement

Question 140

Legal Framework

Answer 140

• Ideas have elements of a public good
• Ideas are non-depletable
• Whether ideas become intellectual property is a complex legal question
• This requires excludability
• And must countries have intellectual property laws to secure exclusion
• The central notion is that of a patent
• Raises complexities internationally
• Patent secured in UK but not in other countries
• Worry innovation can be copied abroad and then imported into UK?

Question 141

A Central Dilemma of Government Intervention

Answer 141

• Protecting innovations give inventors strong incentives to innovate
• They create a monopoly right to use the invention and to generate maximum profits from it
• But monopoly power limit the scope of who can use the technology
• After the event, there would be higher welfare from letting everyone use the technology
• Competition would bid down prices
• But that would undermine investment incentives
• Can’t have both innovation and low prices
• Suggests a trade off between
1. Static efficiency - lower monopoly welfare loss
2. Dynamic efficiency - more innovation

Question 142

A Model of Innovation Model

Answer 142

• An entrepreneur is contemplating putting effort inventing a new good
• Whether they’re successful is uncertain
• If they’re is successful, then they can file a patent
• This will make them a monopolist when selling this new good
• And they can earn monopoly profits
• Is they are unsuccessful then the effort put into it is wasted

Question 143

Basics

Answer 143

• Simple general equilibrium model with two goods
• An old good
• A new good whose price is determined endogenously by setting supply equal to demand
• Everyone consumes the old good and there is demand for the new good when it comes along.
• So we have to specify a utility function for consumers for the new good
• Then derive the demand function
• Then work out the equilibrium price
• Where x2 may or may not exist depending on whether innovation is successful
• Bundle all existing goods into x1
• If there is no innovation, all spending will be on these goods
• The parameter A indexes how valuable the new good is if it comes into being
• Think of it as measuring the significance of the innovation
• The parameter B will govern how price sensitive consumers are when buying it
• If B is high, demon big, fraction small, utility is smaller

Question 144

Basics II

Answer 144

• Consumers have income m comprised of endowments units of x1
• We use the price of good one as the numeraire p1 = 1
• And we do not specify the production process of good 1 as we are not interested in the old economy
• There are M consumers and a single potential innovator
• If they innovate, the innovate charges p for each amount of the new good sold
• Note that we are using p = p2, but there is no risk of confusion since p1 = 1
• We will ignore the role of the innovator as a consumer
• I.e., the innovator only consumes good 1
• They have an endowment of good 1 just like everyone else if they do not become an innovator

Question 145

The Entrepreneur

Answer 145

• Commits effort to innovation
• We suppose that the effort determines the probability of successful innovation
• So let e be the probability of innovating
• And the effort cost of achieving innovation probability e is: c/2(e.)^2
• Work harder = higher probability of successful innovation
• The utility cost of effort is increasing and convex

Question 146

Timing of Events

Answer 146

1. The entrepreneur decides on e which determines the probability that a new good is invented
2. If the good is not invented then it is business as usual in a world of only good 1
3. If the good is invented, the entrepreneur commercialises it and chooses the price at which they will sell it, produces an amount of good 2
4. Whether in situation 2 or 3, production and consumption takes place determining the utilities of consumers and the entrepreneur

Question 147

A World Without Innovation (Stage 2)

Answer 147

• There is only good one
• M + 1 units of the good are produced
• Everyone sets x1 = m and consumes their endowment
• Their utility is also m
• So the old economy is boring and simple

Question 148

Successful Innovation (Stage 3)

Answer 148

• Now good 2 has been invented and the entrepreneur is a monopolist
• Each consumer maximises their utility
• Monopoly Profits at price p are:
Demand x [price - marginal cost]
• Positive demand if can sell above MC
• Otherwise the new good will not “take off”
• So as in a standard monopoly problem price is set above marginal cost
• This means that the entrepreneur makes a profit (due to intellectual property right)
• We can now calculate the profit of the entrepreneur
• We plug in the formula for the optimal price back into the expression for profit
• Profits are higher if:
• The innovation is more useful - higher A - higher monopoly profit
• There is a larger potential market - higher M - more people to sell to
• Consumers are less price sensitive - lower B - charge higher price without discouraging consumption
• So consumers are better off with the new good even though they pay the monopoly price for it

Question 149

The Incentive to Innovate (Stage 1)

Answer 149

• The entrepreneur chooses e in anticipation of the profit that successful innovation brings
• Case for monopoly profit as it generates innovation
• But most efficient way compared to gov or market?

Question 150

The Rate of Innovation

Answer 150

• So innovation is higher is M and A are higher because profits are greater
• More effort for innovations with higher payoff
• Assuming one can get a monopoly right

Question 151

Societal benefits from Innovation

Answer 151

• Will take this to be the sum of profits plus consumer utility
• We want price set equal to marginal cost
• Then profit is 0
• So left to the market, there would be no innovation ???

Question 152

Comparing Optimal Innovation with Market Innovation

Answer 152

• They differ because market innovation depends on big pi and optimal innovation depends on W
• So we need to compare W and big pi
• Suggests innovation rate in the market is too low
1. Then entrepreneur cares only about profits but the innovation also benefits consumers
2. Even if the consumer’s welfare was taken into account, monopoly pricing leads to a welfare loss
3. Inequality created?
*Value of innovation = profit of innovator + utility of consumer, assumes entrepreneurs aren’t altruistic

Question 153

Summary

Answer 153

• The markets will produce innovation by protecting innovator’s property rights
• This creates market power so there is a welfare loss from monopoly and the innovation is not used enough
• The innovator also fails to anticipate consumer benefits over above those that are captured by profits
• So, in principle, the rate of innovation is too low
• There is a prima facile case for government innovation

Question 154

Forms of Government Intervention

Answer 154

• Governments often subsidise private research and development
• They also fund research directly (e.g. through universities)

Question 155

Incentives for Innovators

Answer 155

• The problem with invoking what a social planner would do is that it is not clear how the planner gets innovation done
• Innovation is inherently a problem of private information which research entrepreneurs create using their efforts
• This is why thinkers like Hayek were deeply sceptical that the government could beat the market, individuals don’t want to share ideas with gov
• The essence of effective innovation is the kind of decentralised activity that goes in the market place
• If the effort required to produce innovations is mainly private information, then the government needs to confront this

Question 156

Use of prizes

Answer 156

• The government could offer to pay for a prize to an innovator equal to W
• These have been used in history
• E.g. malaria: public prize better incentive than weak profit incentives provided by the market
• Who would adjudicate if the innovation was successful
• Problem of all the gains from the innovation going to the entrepreneur
• The prize system would create even more inequality in favour innovators than the market!
• So more realistically, society would trade off the size of the reward against the rate of innovation

Question 157

Innovation Incentives

Answer 157

• The effort put in by entrepreneurs is private information so that a social planner cannot contract on effort
• In the world of moral hazard (can’t observe around work going into innovation)
• The innovator has to be given an incentive to innovate
• Under our assumptions that innovators care about profit, this means paying a bonus for innovation (basically a prize model)
• The consumers have to fund these prizes out of taxation so there is a trade-off
• High prizes mean more innovation
• But consumer pay higher taxes

Question 158

The Optimal Prize

Answer 158

• Let b be the prize for innovating
• The entrepreneur will get m if theyre no an entrepreneur
• Society cares about W but has to pay bonuses out of income
• Lump-sum tax assumed
• If there is innovation, society will set price equal to marginal cost
• So optimal incentives do not beat the market in terms of the rate of innovation
• This is general that since p = MC, welfare is higher with government ownership of the innovation if it does not use its monopoly power• One critique of the market is that a lot of innovation is about creating things that we do not need

Question 159

Overconfidence

Answer 159

• Quite a few psychological studies find that entrepreneurs appear to be overconfident
• For example, exaggerating the success of probability
• This will boost the rate of innovation
• An individual could choose to become an entrepreneur even if their “objective” outside option is lower
• But society could benefit

Question 160

Intrinsic Motivation

Answer 160

• Non-pecuniary factors could be important
• Cared about idea for idea’s sake
• Social entrepreneurship = social innovation
• Requires socially responsible customer base

Question 161

Incorporating Intrinsic Motivation

Answer 161

• Then (rather trivially) that would mean that their choice of e would be higher
• And individuals with mu > 0 are also more likely to become entrepreneurs
• This may matter a lot when there is the possibility social entrepreneurship
• Products that have a “good cause” associated with them
• Then (rather trivially) that would mean that their choice of e would be higher
• And individuals with mu > 0 are also more likely to become entrepreneurs
• This may matter a lot when there is the possibility social entrepreneurship
• Products that have a “good cause” associated with them