1 Introduction to Public Economics 1 ULB Prof. A. Estache Lecture 4: Public, Common and Club Goods + Cost Benefit Analysis.

1

Introduction to Public Economics 1ULB

Prof. A. Estache

Lecture 4:

Public, Common and Club Goods +

Cost Benefit Analysis

Where this lecture fits in the course

In lecture 1, I reminded you about market failuresi.e. When the market does not lead to an efficient

outcome (and by now, you need to be able to define what efficient means...)

So what leads to market failures?• Externalities (lecture 3)• Public goods + Clubs Goods + Common Pools

(lecture 4=today)• Information asymmetries (lecture 5)• Imperfect competition (a bit in lecture 7... If not

enough for you, join me for the course on regulation in MA...)

2

3

Coverage today

1.Public Goods

2.Common Goods

3.Club Goods

4.Cost Benefit Analysis

4

Characteristics of Goods (1)

Goods can be classified according to two attributes:

whether they are excludable and

whether they are rival in consumption

What does that mean????

5

Characteristics of Goods (2)A good is excludable if the supplier of that good can prevent people who do not pay from using it.

=> non-excludable means that individuals cannot deny each other the opportunity to consume a good.

What does it mean for cost recovery?

A good is rival in consumption if the same unit of the good cannot be consumed by more than one person at the same time.

A good is non-rival in consumption if one individual’s consumption of a good does not affect another’s opportunity to consume the good.

What does it mean for pricing?

6

Characteristics of Goods (3)

Since goods can be classified according to how they fare in terms of these characteristics, all goods can be classified into 1 of 4 types of goods, each defined in terms of excludability and rivalry:

Private goods, which are excludable and rival in consumption, like wheat, cars, laptops

Public goods, which are nonexcludable and nonrival in consumption, like a public sewer system or defense

Common resources, which are nonexcludable but rival in consumption, like clean water in a river

Club goods (=artificially scarce goods) , which are excludable but nonrival in consumption, like pay-per-view movies on cable TV, computer software, toll roads

7

Characteristics of Goods (4)

Rival in consumption

Non rival in consumption

(=> efficient price is 0!)

Excludable

(=> Can charge!!!!)

PRIVATE GOODS•Wheat, bread, cars, phones, CDs,….

CLUB GOODS•Computer software•Toll roads (uncongested)•Cable TV

Non Excludable

COMMON RESOURCES•Sidewalks•Clean water•Biodiversity

PUBLIC GOODS•National Defense•Street lights•Public sanitation

8

• Imagine 2 individuals, Ben and Jerry• They need to decide between consuming 2 pure private

goods• Look at each market separately first• Find the optimal level of production given the demand

by Ben and then by Jerry• The most obvious result for a private good is that consumers

demand different quantities of the good at the same market price=> both can consume the good according to their taste…even if this taste is different• For every price of ice cream, we compute the demand by Ben

and by Jerry and then add them up as total market demand

Optimal Provision of Private Goods

9

Optimal Provision of Private Goods (add up horizontally)

10

Optimal Provision of Private Goods

What does it all mean analytically?

Imagine 2 consumers have a different taste for Cookies (c) and Ice Cream (ic) => the optimality condition for the

consumption of private goods is:

In equilibrium, MRS = MC

And since without market failure, MRS=MSB and MC=MSC

Private solution leads to efficient outcome where MSB=MSC

Since on the supply side, equilibrium requires: MCic = Pic and MCc = Pc

Set Pc as the numeraire (i.e.=1), from (1) , MRS = Pic

11

NOW:…Optimal Provision of Public Goods• Now imagine that Ben and Jerry chose between a missile

and cookies…(1 public good vs 1 private good)• Set again the price of cookies to 1• Major difference with private good case is that

consumers cannot tailor their consumption of missiles to their taste!

• => this forces to look at the market vertically in the graph• => Add up the price that each person is willing to

pay for a given market quantity• Look at number of missiles chosen and see how much

each is willing to pay• Ben is willing to pay 2 for the 1st missile and 1 for the 5th

• Jerry is willing to pay 4 for the 1st and 2 for the 5th• while equilibrium price is 3…for 5 missiles (SMC=SMB)

12

Optimal Provision of Public Goods (add up vertically)

13

Optimal Provision of Public Goods

What does this all mean analytically?

•The missile is worth MRSBm,c to Ben and MRSJ

m,c to Jerry

so the total value to society is MRSBm,c + MRSJ

m,c .

•The Social Marginal Cost is the same as usual (MC of producing a missile)

•The social-efficiency-maximizing condition for the public good is

Social efficiency is maximized when the marginal cost is set equal to the sum of the MRSs, rather than being set equal to each individual’s MRS.

14

Summing up…• For private goods, socially optimal for firms to

produce until MC=MB to the marginal consumer (private competitive outcome)

• For public goods socially optimal to produce until MC is = the benefit of ALL consumers combined => risk of free riding since some may not want to pay => risk of market failure since competition will NOT deliver the good if consumers do not pay for its production• This is because the private good is RIVAL (once

consumed by 1 user, it is gone)• Since the public good is NON-RIVAL, it can be consumed

jointly by all users, society wants producer to account for the sum of all user’s preferences!

15

So how do you provide public goods???

• Private vs. public provision given the risk of free riding????

• HUGE DEBATE…

• but useful to study the two options analytically before looking at what we know about the evidence….in a later lecture

16

What’s key part of problem of providing Public Goods?... hard to recover its cost!

If get the benefit of the good without paying…will try not to pay!

so free riding is a serious issue!

a private firm is unlikely to be interested in providing the service…This is a primary justification for the existence of government.

17

A closer look at the problem with the Private Provision of PGs

Private-Sector Underprovision

Free rider problem…more formally

When an investment has:1. a personal cost…2. but a common benefit, =>No individual has an incentive to pay for providing the efficient

quantity of a public good because each individual’s marginal benefit is less than the marginal social benefit

Individuals will pay less than they should to finance the investment!

=> Hard to recover the costs of provision

18

Think of the public good as a special externality

• When 2 users benefit greatly from the production of a public good, then each contribution of each user benefits the other

• As we saw in the discussion of positive externalities…

• This leads to underproduction!

• =>Case for government intervention in the delivery of public goods from a socially optimal viewpoint as in the case of externalities…but forms of intervention may be different…including actual delivery…• Major historical justification for public

enterprises, public education, public health…

19

Example: Global Public Goods

International public goods, global and regional, address issues that:

• are deemed to be important to the international community, to both developed and developing countries; 

• typically cannot, or will not, be adequately addressed by individual countries or entities acting alone, and, in such cases;

• are best addressed collectively on a multilateral basis.

20

Why should we care about GPG?Why should we care about GPG?

It’s about the world wide impact of “Globalization”.

Before After

HIV Infections Worldwide

0 (1960) 40 m (2005)

Carbon Dioxide Concentrations

300 ppm (1992) 380 ppm (2005)

Refugees Worldwide 5.7 m (1978) 19.2 m (2005)

Avian Flu Negligible (2005) ? (2010)

21

Main issues with GPGs: UNDERPROVISION

• Institutional arrangements are unclear;

• Cross-border externalities can discourage provision of GPGs;

• Policy making is typically at national level; International cooperation is difficult to negotiate and implement.

22

So where does the underprovision come from?• Cross-border spillovers can create a financing

gap;• International Cooperation is needed for a solution • To shore up GPG financing, part of Official

Development Assistance (ODA) is committed to GPGs (up to 25% by some estimates);

• In key areas, rich nations finance provision of GPGs • E.g. Global Fund to Fight Aids, TB, and Malaria

and Global Environment Fund• However, concerns are raised about resources

flowing from activities reserved for poor countries to activities that benefit all countries including rich ones.

23

The Free Rider Problem in Practice A P P L I C A T I O N

…but keep in mind that evidence suggests free provision is not necessarily a solution to free riding…

Cambridge, England, tried to provide 350 free green bicycles scattered throughout the city.

Users were expected to return each bicycle to one of 15 stands after its use.

Within four days, not a single bicycle could be found, most having been likely stolen and repainted.

24

So … how to deal with the free rider problem?

Private vs. public solutions…again!!!

=> …useful to know when the private sector should be able to deliver:

1. When some individuals care more than others

2. When altruism prevails3. When “warm glow” happens

25

When Is Private Provision Likely to Overcome

the Free Rider Problem?1. Some individuals care more than others

• Some individuals may have an especially high demand for the public good (heterogeneity of consumers)

• => these have a higher willingness to pay which may offset the underinvestment by others

• The same may happen if some individuals have higher income even if same taste• Helps but does not deal fully with the free

rider problem => still may have underprovision

26

When Is Private Provision Likely to Overcome

the Free Rider Problem?

2. Altruism• When individuals value the benefits and costs to others

in making their consumption choices• => may be willing to contribute more to a public good

even free rider problem suggests they should not• Evidence:

• from laboratory experiments suggests that 30-0% of participants are willing to contribute to a public good (never 0!)

–=> are we more altruist than we thought?• The explanation is social capital: The value of altruistic and

communal behavior in society.–More altruist when trust more the others–Reciprocity and sharing matters

»Bangladesh private trash collection example» In practice, not enough of it

27

When Is Private Provision Likely to Overcome

the Free Rider Problem?

3. Warm glow model• Model of public goods provision in which

individuals care about both the total amount of the public good and their own particular contributions as well

• Some people like to get a medal, a statue, a park with their name in exchange for contribution

• => the public good can become like a private good…but still not fully private since does not account for the benefits to others of the public good provision

28

• ...it looks like the private sector will usually underprovide the public goods …

role for government to intervene1. By providing itself2. By mandating private actors to provide

• However, governments have their own share of problems when trying to provide public goods1. Crowding out2. Difficulty of measuring costs and benefits of

public goods3. Difficulty of determining preferences for public

goods

NOW: What about the Public Provision of Public Goods?

29

1.The problem of crowding out

crowding-out?As the government provides more of a public good, the private sector provides less

Comments…• Classical example of unintended consequence of

government action• Risks of lack of robustness of economic equilibria• Usually partial rather than total

30

When can full Crowd-Out or a partial Crowd-out happen?

There may not be full crowd-out if I care about my own contributions. If I get utility from my particular contributions for any reason, then an increase in government contributions will not fully crowd out my giving.

PARTIAL: Warm Glow

Evidence on Crowd-Out?Unfortunately, the existing evidence on crowd-out is quite mixed. While there is no evidence for full crowd-out, there is also no consensus on the size of this important individual response to government intervention.

Suppose that some people contribute more for PG than others, either because they are richer or because they have a stronger preference for the public good. => If forces everyone to contribute, the gvt increases total PG provision but people will no longer buy from the private sector

FULL: When taxing non-Contributors to finance PG

31

2. Difficulty of measuring the Costs and Benefits of Public Goods

Up to now, we assumed that government had full information on costs and benefits…

but if not…measuring costs and benefits can be complicated.

• With uncertainty, risk of free rider problem increases or changes in nature if leave it to the private sector (in particular with respect to the demand side…and hence benefits and willingness to pay)…

• does the private company deliver what it is supposed to deliver if we can’t measure some of the costs and benefits?

• Moreover, financial cost-benefit of private provider may not match the social cost benefit analysis of a public provider

32

Illustrating the difficulty of measuring the Costs and Benefits of Public Goods

• Consider the case of an highway projects…

• What if, without this highway project, half of the workers on the project would be unemployed?

• How can the government take into account that it is not only paying wages but also providing a new job opportunity for these workers?

• What is the value of the time saved for commuters due to reduced traffic jams?

• And what is the value to society of the reduced number of deaths if the highway is improved?

33

3. How Can We Measure Preferences for the Public Good?

3 major problems when trying to figure what is it that people want when deciding about a public good

1. Preference revelation: individuals may not be willing to tell the government their true valuation because the government might charge them more for the good if they say that they value it highly.

2. Preference knowledge: even if individuals are willing to be honest about their valuation of a public good, they may not know what their valuation is, since they have little experience pricing public goods such as highways or national defense.

3. Preference aggregation: how can the government effectively put together the preferences of millions of citizens in order to decide on the value of a public project?

These difficult problems are addressed by the field of political economy, the study of how governments go about making public policy decisions, such as the appropriate level of public goods.

34

...hmmmmm....Do you still remember the various types of goods?

Rival in consumption

Non rival in consumption

Excludable PRIVATE GOODS•Wheat•Bread, cars, phones, CDs

CLUB GOODS•Computer software•Toll roads

Non Excludable

COMMON RESOURCES•Clean water•Biodiversity

PUBLIC GOODS•Defense•Public sanitation

35

Now ….A few words about the problem of Common Resources

A common resource is nonexcludable BUT rival in consumption:

=> you can’t stop me from consuming the good, BUT more consumption by me means less of the good available for you.

Examples: clean air and water as well as the diversity of animal and plant species on the planet (biodiversity).

In each of these cases the fact that the good, though rival in consumption, is nonexcludable poses a serious problem.

36

The Problem of Overuse of Common ResourcesCommon resources left to the free market suffer from overuse:

a user depletes the amount of the common resource available to others BUT does not take this cost into account when deciding how much to use the common resource.

For a common resource, the marginal social cost of my use of that resource is higher than my private marginal cost (the cost to me of using an additional unit of the good)

The following figure illustrates this point…

A Common Resource

The supply curve S, which shows the marginal cost of production of the entire fishing industry, is composed of the individual supply curves of the individual fishermen. But each fisherman’s individual marginal cost does not include the cost that his or her actions impose on others: the depletion of the common resource. As a result, the marginal social cost curve, MSC, lies above the supply curve; in an unregulated market, the quantity of the common resource used, QMKT, exceeds the efficient quantity of use, QOPT.

38

The Efficient Use and Maintenance of a Common Resource

To ensure efficient use of a common resource, society must find a way of getting individual users of the resource to take into account the costs they impose on other users.

Like negative externalities, a common resource can be efficiently managed:

1. by Pigouvian taxes (tax or otherwise regulate the use of the common resource)

2. by making it excludable and assigning property rights,

3. or by the creation of a system of tradable licenses for the right to use the common resource.

39

A relevant digression on congestion pricing Some externalities can be thought of as common good problems …Look at

airports use, …anyone is entitled to use the airport…airport owner cannot exclude anyone…but airport turns out to be rival in consumption in case of congestion …so useful insights to deal with common goods as well

When demand for this common good exceeds capacity, today….users don’t pay for the delay costs imposed on others

Solution? Charge users for the costs they impose, and delays will fall and social welfare will improve

There is an optimal congestion price for the use of that common resource that can be charged that maximizes social welfare

Optimal Congestion Fee PT

PP

QTQO QP

MB

Private MCSocial MC

Passenger Trips

$

40

How do you do all this in practice?Set a target level of delay or operations at the airport

An optimal level (which varies with mix of aircraft), or A physical target

Establish a market mechanism that facilitates moving toward the target An auction for time slots…so different prices for different slots, so

different ticket prices…

Be prepared to have prices vary across small units of time—e.g., 15 minutes

Everyone pays the same price in the same time period

Establish a way to reset prices as demand changes As often as necessary to stay close to target

Holiday vs non holiday prices Keep users informed

Could conceive similar rules for uses of common resources to avoid abuses

41

..and now a few words about

Club Goods (Artificially Scarce Goods)

Rival in consumption

Non rival in consumption

Excludable PRIVATE GOODS•Wheat•Bread, cars, phones, CDs,

CLUB GOODS•Computer software•Toll roads

Non Excludable

COMMON RESOURCES•Clean water•Biodiversity

PUBLIC GOODS•Defense•Public sanitation

42

… so how do we deal with Club Goods??? (Artificially Scarce Goods)An artificially scarce good is excludable BUT nonrival in consumption.

Because the good is nonrival in consumption, the efficient price to consumers is zero.

However, because it is excludablesellers can and do charge a positive price, which leads to inefficiently low consumption.

It is made artificially scarce because producers charge a positive price but the marginal cost of allowing one more person to consume the good is zero.

The problems of artificially scarce goods are similar to those posed by a natural monopoly.

Useful example to think of for a government: Toll roads

Club Goods

In this example the market price is $4 and the quantity demanded in an unregulated market is QMKT. But the efficient level of consumption is QOPT, the quantity demanded when the price is zero. The efficient quantity, QOPT, exceeds the quantity demanded in an unregulated market, QMKT. The shaded area represents the loss in total surplus from charging a price of $4.

44

Conclusions on non private goods…1. A major function of governments at all levels is the provision of public goods… directly or

indirectly2. In some cases, the private sector can provide public goods, but in general it will not achieve the

optimal level of provision.

3. When there are problems with private market provision of public goods, government intervention can potentially increase efficiency even if the scope for private financing continues to be an option when only part of the market characteristics fail (club goods)

4. Whether that potential will be achieved is a function of both the ability of the government to appropriately measure the costs and benefits of public projects and the ability of the government to carry out the socially efficient decision.

45

So how do we decide whether do take on a project to deliver a public good or not????

That’s what Cost-Benefit Analysis is all about!

It translates many of the hard concepts you have had to learn over the last 2 years…into HARD numbers you can use to make decisions!i.e. how do we achieve Pareto improvements

through specific project?

46

…First, measure the Costs of Public Projects

Example: Consider the renovation of the intercity highway system.

• It has 3 sources of costs: Asphalt, Labor and Maintenance…

• It has 2 major benefits: reduced travel time and fewer fatalities

The goal of CBA is to quantify these costs and benefits!(note you could also have to account for environmental cost, visual costs, …)

47

Measuring the Costs of Public Projects?

1.Current vs. opportunity cost?

2.Perfect vs. imperfect markets?

3.Future markets?

48

Measuring the Costs of Public Projects?

1. Current costs vs. opportunity costs(financial vs. economic view of the world!)

• Current costs (CC): • Accounting method that calculates costs solely by adding up what the

government pays for inputs to a project, – (calculates benefits solely by adding up income or government revenues

generated by the project)• cash flow accounting….used by accountants…but different from SMC

concept we used so far….• Opportunity cost (OC)

• It is the SMC (social marginal cost) of any resource concept we have used up to now:

• In practice, it is the value of a resource in its next best use

The cost to society of using any input is not its Current Cost by its Opportunity Cost

49

What do these basic cost concepts mean in practice?

2. Perfect vs. imperfect markets for inputs• If a good or a service (say labor) is sold in a competitive

market, its OC=CC• Example: Asphalt, next best use is to sell it to someone else

=> value is its market price, so in this case, OC=CC• If a market is distorted, this no longer holds!

REMEMBER: Economic costs are only those costs associated with diverting the resource from its next best use..so key to distiguish between financial costs and economic costs and recognize that

difference between the two is simply a transfer from one agent to another….NOT an economic cost!

50

What do these basic cost concepts mean in practice?

Example 1: imperfect labor market=> unemployment utility drawn from having more leisure matters, so working has an

opportunity cost = to the value of leisure to the unemployed If this opportunity cost is higher than wage, unemployed is not interested

in working Say, wage is 20/hour and leisure is valued at 10/hour and half of the

people we need to do 1 million hours of work are employed and the other half is unemployed

the opportunity cost of hiring 1000 days of work is 20*.5m + 10*.5m = 15 million, even if the gvt is actually paying 20 million in cash!

The cash cost to gvt has 2 components: OC of labor + transfer of rents (rent = payments to the resource deliverer (worker) beyond those needed to get the resource, i.e. could have gotten the jobless for less)

Rent is NOT a cost to society, it is just a transfer from gvt to unemployed Rent included in cash accounting but NOT in economic analysis of costs

Example 2: if asphalt sold by a monopoly => p>MC

=> OC = MC and P-MC = transfer

51

3. Measuring Future Costs…• Not all costs take place instantaneously!

• Look at maintenance…it involved labor and material over the years to keep the highway in good shape

• So we need to be able to have a system that accounts for both “one time immediate” costs (construction) with future streams of costs that spread over the years (maintenance)

• This is why we rely on Present Discounted value or PDV (read pp102-104 in Gruber)

• The value of each period’s dollar amount in today’s terms• A $ tomorrow is worth less than a $ today since if we have it today, we can earn interest

rates on it at the private level, or the economy can make a social return on it• Mathematically, for a discount factor r (i.e. opportunity cost to firm of making the

investment--i.e. next net of tax return on best alternative investment possible– for a private projects or the social discount rate—i.e. gross of taxes return of the best private investment possible—for a social/gvt project) and for a payment in each future period are F1, F2, . . . and so on, then the PDV is computed as

• …so future maintenance cost in this example…need to be discounted to compare them to today’s construction costs

• For a social discount rate of 7%, the 10 million future stream of maintenance costs has a PDV of 143 million (=10/0.07)

52

Summary of costs for now TOTAL COST in today $ is 100 (asphalt)+15 (labor) + 143 (maintenance)=258 m.

53

Measuring the benefits????• Benefits are much harder to measure than costs!• Why? Because harder to use market values to place a value on

benefits!• Example: how do we value the time saved by a transport project

such as our highway project?• For producers, it cut the transport cost of delivering products

=> more supply (rihgt shift in Supply!)=> can measure the social surplus achieved = Social Benefit

• For consumers…much harder to estimate the value of time saved

• Economists have several techniques to do it• ...none of them is perfect!!!!

• The options are:–Market-based measure of value of time = Wages!

–Survey-based measures = Contingent Valuation–Behavior-based measures = Revealed Preferences

54

Using Market-Based Measures to Value Time:

WagesSuppose we can show that the time that individuals save from driving faster is spent at work

=> approximating the benefit through wages is OK

This theoretical proposition runs into some problems in practice:

What if time saved is partly spent working and partly on leisure? No problem because OC of leisure is wage…

What if labor market imperfections?...more problems

Individuals can’t freely trade off leisure and hours of work; jobs may come with hours restrictions, so not free to use the saved time as workers wants it => overestimates the benefits

There may be nonmonetary aspects of the job…=> even more problems since wage underestimates the benefits

Economists have had to develop alternatives

55

Using Survey-Based Measures to Value Time:

Contingent Valuation

•contingent valuation? Asking individuals to value an option they are not now choosing or do not have the opportunity to choose, then aggregate individual valuations

• i.e. how much would you pay to save 5 minutes on your drive?

•Often the only way to assess benefit of a public good!• i.e. how much do you value saving a rare bird?

•Serious Problems However…• Isolation of issues matters (answers differ when 1 question is

asked alone or when part of larger set)• Order of issues matter (asymmetric…psy?)• Embedding effect matters (some questions are perceived as

wider than they really are)

56

Using Revealed Preference to Value Time

•revealed preference ? Letting the actions of individuals reveal their valuation.

• non economist would ask people how do you value your own time?• Economist would read into the people’s behavior what they reveal about

their value of time• People may lie but their actions, since they reflect an optimization process do

not!•Example, compare 2 houses otherwise similar, but one is closer to your job…the difference in price you are willing to pay for the closest house reveals your valuation of time!• = a market based valuation•Problems?

• Hard to find natural experiments that work (similar houses?)• Multiplicity of dimensions that drive decisions makes it hard to isolate 1

dimension (e.g. time in our case study)• Ideally, run a controlled experiment…but not easy…but possible

• Hedonic market analysis can work for many issues• => risk of underestimating or overestimating the true value

57

What if the social benefits involve saving lifes? How do you measure the value of a life?????

Valuing human lives is the single most difficult issue in CBA.

Real life examples range from $31.25 million (GM killer trucks) up to $300 million in the London commuter train crash

Many would say that human life is priceless, that we should pay any amount of money to save a life. By this argument, valuing life is a reprehensible activity; there is no way to put a value on such a precious commodity…so all gvt expenditures should be allocated to saving life

We could claim that virtually any government expenditure has some odds of saving a life.

To escape the impotence that would be imposed by the “life is priceless” argument, one needs to be able to place some value on a human life.

How so?

• Market-based measure of value of time = Wages!• Survey-based measures = Contingent Valuation• Behavior-based measures = Revealed Preferences• Government Revealed Preferences…

58

Valuing Saved Lives? Check the 4 approaches

1. Using Wages to Value a Life

2. Contingent Valuation

As with valuing time, the market-based approach to valuing lives is to use wages: life’s value is the present discounted value of the lifetime stream of earnings … but what about value of time not worked?.

Ask individuals what their lives are worth.i.e. differences in payment with varying distance to a nuclear power

station with light radiations which would reduce their life by 1 year?•3. Revealed Preference

As with valuing time savings, the method preferred by economists for valuing life is to use revealed preferences. We can value life by estimating how much individuals are willing to pay for something that reduces their odds of dying.

ie. How much are drivers willing to pay for a passenger air bag? $350 + since odds of it saving a life is 1/10000=> life is worth 3.5 million at least

i.e. 2. compensating differentials: Additional (or reduced) wage payments to workers to compensate them for the negative (or positive) amenities of a job, such as increased risk of mortality (or a nicer office).

59

4. Government Revealed Preference

60

...REALLY SERIOUS ISSUE

Discounting The Future

A particularly thorny issue for cost-benefit analysis is that many projects have costs that are mostly immediate and benefits that are mostly long-term.

Example: global warming…you pay today for benefits for you children…so how do you discount (STERN REPORT)

Should the benefits of long lived projects be treated differently for current and future generations?

61

Putting It All Together

62

A few last words of caution on CBA3 common challenges

1. Common Counting MistakesWhen analyzing costs and benefits, a number of common mistakes arise, such as:

Counting secondary benefits. New road make create new local business…but reduce business elsewhere =>

what matters to assess benefits is NET creation of surplus Counting labor as a benefit.

Labor is part of costs from the viewpoint of the project…not the benefits Double-counting benefits.

Do not include increase in price of house due to benefit of saving time!

2. Distributional Concerns

3. Uncertainty

The costs and benefits of a public project do not necessarily accrue to the same individuals.

For instance, commuters benefit from faster time, but non-drivers residents have more noise and traffic to deal with.

In theory, losers could be compensated by winners so all win…but seldom happens… HENCE, may need to have distributional weights for rich and poor….but controversy!

The costs and benefits of public projects are often highly uncertain Should gvt prefer certain but less redistributive benefits to potentially high but

uncertain benefits?

63

What happens if really can’t assess benefits?

Rely on cost effectiveness analysis!

Cost-effectiveness analysis? Compute only the costs of the various projects considered and choose cheapest for a given public good output we are interested in

64

Conclusion•Government analysts at all levels face a major challenge in attempting to turn the abstract notions of social costs and benefits into practical implications for public project choice.

•What at first seems to be a simple accounting exercise becomes quite complicated when resources cannot be valued in competitive markets.

•Nevertheless, economists have developed a set of tools that can take analysts a long way toward a complete accounting of the costs and benefits of public projects.