Lecture Seven: A Closer Look At Public Goods
Sources: Various, but includes Public Finance (Ulbrich) and Public Finance and The American Economy (Bruce).
Remember our definition:
Nonexcludable (nonexclusive) – cannot be held from any member of the group once it is supplied to the group.
Nonrival – one person’s consumption of a good does not affect another’s consumption. The benefit of a public good is shared by all members of the group receiving the good regardless of whether the individual members have chosen to pay for it.
Pure Public Good (marginal cost of an extra user is zero) -
Pure Private Good -
Do public goods exist? Technology matters!
Mixed Public Goods: Since it is hard to find a true public good – NOW the new definition includes any good that is either nonrival (but doesn’t have to be nonexcludable) or nonexcludable (but doesn’t have to be nonrival). More in detail later.
National defense if often seen as nonexcludable (in some cases). If Betty is defended by Bob because he kills Bill – then Barb is also defended (not excluded since Bill won’t kill her now either) and nonrival (Betty’s consumption of the defense does not preclude Barb from consuming it as well).
In some cases, however, Bob could defend Betty but let Barb get killed – it is situational.
We will discuss other examples later.
Governments supply private goods and markets supply goods that some would consider public goods. So is the economic theory breaking down?
So the theory goes (with pure public goods). . . Market Failure With Public Goods
So why won’t markets supply the public good?
With a private good, the efficient quantity of a private good can be achieved by charging a price to ration the quantity available. Consumers will buy as long as the MWTP is at least as high as the price.
But if a good is nonrival and nonexcludable – how can we charge a price for it? Who would pay?
Nonexcludability and the Free Rider Problem:
Free riders:
If consumers would benefit from the public good, why don’t they simply provide the good? After all, everyone can be made better off with a more efficient allocation of resources. Why don’t we simply agree to do our part by providing the public good?
Standard economic explanation: HANDOUT (IN Class Exercise)- Public Goods and the Prisoner's Dilemma
“Efficient” Output of a Public Good:
Main problem: How much public good should be produced? We don’t have prices to guide us.
Enjoying more public goods (like national defense) means that we must enjoy fewer private goods like food and clothing.
What criteria will we use for “efficient”?
Let's try "allocative efficiency" -- or where the marginal willingness to pay = the price (basically where the marginal benefit = marginal cost) or where S = D.
GRAPH:
HANDOUT (In Class Exercise) - Example: Planning a party.
It's nice in theory, but what about in practice? Markets try to find this efficient allocation of resources. Of course we would have to have perfect information (and a static situation) for this to happen. We never really know if we are perfectly “allocatively efficient” – plus, even if it happened, the situation would change momentarily. However, market prices, profits and losses provide us with a proxy of value – that gives us some idea as to how resources should be used – and their opportunity cost.
Can government planners ever find it – or know that they are using resources in an allocatively efficient way? Economists would say no. Resources move towards the value of the planners – not allocated towards the value of consumers.
Some might prefer this – thinking that planners know better how to use resources than consumers.
But even so, is it possible for planners to move resources such that even the basic necessities of life are provided. Evidence indicates that this is very difficult – if not impossible (the collapse of planned economies around the world). Without knowing how to use a resource – where it would be used in a more valuable way – they can only guess. Causing shortages and surpluses that linger.....
Comparing Efficiency Conditions for Public and Private Goods
Efficiency conditions for nonrival and rival goods are different. Indeed, they are reserved, in a sense.
Note that with the rival good, wine – we determined the point where the MWTP is the same for everyone (where = MC) at $1.00 and then summed up the glasses of wine.
People with different tastes consume different quantities. We can each consume up to the point where the MC = MB and markets are used to find this point. Markets try to find the prices (MC) that equate MB to consumers or where S = D. They aren’t perfect but at least the actors in markets are trying to find this point…
For the nonrival good, songs – the quantity is the same for everyone (everyone listens to the song at the same time) and then we summed up the MWTPs.
People with different tastes have different MWTPs. So the problem is – in reality we can’t sum these different MWTPs up because we can’t measure them. Therefore, each person wants a different amount of the public good but they all get the same amount.
Some people might have negative MWTPs for some public goods, such as national defense. That’s easy to handle in our little example, just subtract out the negative amount – but not so easy to handle in reality.
How do we find the point where the sum of the MWTPs is equal to the marginal cost of defense, for example?
The Private Provision of Public Goods
Go back to our Free Rider problem - In a small group like this though, do you think George and Martha could strike a deal and both stand guard? Probably.
The “cooperative outcome” is possible – especially when we can easily determine in a small group if someone is free riding.
Roads (often seen as public goods – although they really don’t meet our definition). They are excludable and, if crowded, rival. If they are not crowded, then they are considered nonrival (and hence a public good under our new definition).
Roads were provided privately through cooperative efforts before the government moved in to provide them.
Another example – lighthouses.
But what about a big group? Seems plausible if some benefited enough to pay for the public good even when others are free riding. But then would the right quantity be provided?
The typical theory says, however, that the larger the group, the more severe is the potential free rider problem, and hence the more likely it is that a public good could not be financed by voluntary contributions.
Even if one person doesn’t contribute towards the production of the good, other people may finance the project – but if enough people thought this way, the good would not get produced.
It all depends upon:
1.
2.
3.
Note that smaller communities manage to provide fire protection (seen as nonexcludable – if a fireman puts the fire out on your neighbors house, your house is protected) through volunteer fire departments.
Therefore we do have Problems with the Government Providing Public Goods:
1. Preference Revelation Problem –
In fact some people might have an incentive to misrepresent their MWTP schedules to the government.
Example: Let’s say that the government levies a tax in order to pay for the public good. George thinks that the amount of tax he pays is fixed and is not affected by what he tells the government, while the amount of the public good he receives is affected, he has an incentive to overstate his MWTP for the public good.
He believes his MC of the public good is zero, he prefers the quantity where his MWTP = zero as well.
On the other hand, if Martha thinks the amount of the tax she pays depends on the MWTP schedule she reveals, but that the quantity of public goods she receives is the same in any event, she has an incentive to understate her MWTP.
This is just the free-rider problem. By understating her MWTP, Martha thinks that she will get the public good but that the taxes will be paid by someone else.
2. Burden sharing –
With markets, the cost of the good is borne by those that benefit from it. But the government can’t do this unless it knows individual MWTP schedules.
So it must use another criterion, such as ability-to-pay. The government simply observes different people’s ability to pay in terms of their income or wealth. This is the most common way of sharing the burden of the public good. In other words, national defense is funded through a progressive income tax.
Mixed Public Goods:
Three cases:
1. a nonrival good that is excludable so a price can be charged,
2. a congestible public good that is excludable,
3. local public goods
Excludable Public Goods: excludable but nonrival - Example:
These can be provided privately because you can exclude people and charge a price.
Congestible Public Goods: a good that is nonrival when there are not too many users, but the marginal cost of an extra consumer rises when the number of consumers becomes too large. This is called a congestive cost.
Example:
Club Goods:
Examples:
These are provided for different reasons than the fact that they are considered mixed public goods since they are easily provided by private firms.
Say, for example a public park (or even a national park - like Yellowstone). If it does get congested, who should be let in and who should not be? Plus, what is the "best" use of the land, it's resources, etc. There's no way of obtaining individual values.
A possible solution: private property rights -- the parks bought by conservation groups who have their own values... See reading by Dwight Lee.
Lee claims the values can be identified AND people who have very different values can peacefully coexist.
Problems?
Local Public Good: Available only to consumers who live within (or visit) a limited geographic area.
Example:
Local public goods are usually provided by local governments. There is a geographic dimension.
Some public goods have mixed properties, being both a local and national public good at the same time. A section of an interstate highway, for example is mostly a local public good but also benefits travelers who are just passing through. Also benefits people living anywhere in the country who use goods from trucks who use the highway.
Local public goods are usually provided by local governments.
Jurisdictional Spillovers:
Example:
Local governments have an incentive to free ride in providing public goods with spillovers. Sometimes cooperative agreements among local governments are made – or a higher level of government gets involved.
Applications of Public Goods Theory
Defense Spending in the United States:
Are we spending too much or too little? Can’t answer that question since we can’t add up the MWTP of all citizens in order to equate it to the marginal cost of defense.
The only way to express our preference for or against national defense is at the ballot box – and in publications, etc.
There is certainly a case to be made that national defense is a transnational public good. Defense spending by the U.S. benefits other countries (allies) (if one country is protected - does that also protect other countries as well (non-rival, non-excludable?).
Is there then free riding going on? Probably. Maybe some countries free ride off of other countries that provide defense against potential predators of more than one country (which could be a reason why some countries spend very little on national defense).
Some might argue that national defense is a transnational public bad? Is there such a thing?
Knowledge Capital: intellectual property rights. Knowledge is a non-rival good – so it is a mixed public good. Also, once it is supplied, it is often hard to exclude others from it.
The answer: patents and copyrights.
If a farmer invents a process that increases the crop yield of his or her land, that knowledge can be used by other farmers at no further cost to society. However, it is not in the interest of the first farmer to share the knowledge and lose his competitive edge. If all farmers increase their yields, this drives down prices and the benefit of the invention is enjoyed by consumers, not farmers. There would be little or no incentive to produce the invention.
In order to maintain the incentive, the individual farmer must benefit from the invention by keeping it a secret. But, as the theory goes, providing the information to the economy would be more beneficial because it is a non-rival good.
So supposedly there is a market failure.
Answer is patents. For 17 years. In order to receive the patent, the inventor must make the knowledge public so others can build upon it.
Also, the government often finances R&D with taxpayer money. Grants, etc. for basic research. The knowledge doesn’t belong to anyone and becomes public.
BUT, the problem again is knowing how much to spend and where to spend it. Competing interest groups will want their “pet research project” funded – how do we know which one to fund? There’s no trial and error process that gives us the value of these projects.
Roads – Mixed public good (congestible public good):
Roads were most often privately funded and built prior to the government getting involved (which started mostly with the Federal Road Aid Act in 1916 and then federal funding grew substantially during the FDR years, National Recovery Act.)
"Shares" were sold to those who wanted the roads built. Although most roads did not make money for the shareholders, they wanted the roads to be built (and therefore contributed) because of the benefits they received from having the roads -- it made the overall payoff greater than the cost. (see Dan Klein, in the The Voluntary City).
So the idea that they would not be provided by the private sector does not match history.
Car manufacturers have been a very powerful lobby in promoted federal funding of roads. Why?
Two controversial issues with the interstate highway system:
1. the decision to route the highways through densely populated urban areas rather than around them. Increased the cost -- the federal government financing local roads.
2. no tolls charged. The marginal cost of another user of a road is not zero -- especially once the road gets congested but even without congestion. Wear and tear, etc. So charging tolls makes sense to the marginal user (congestion costs and wear and tear costs).
Standard economic theory suggests then: figuring out the marginal cost of the vehicle (congestion and wear/tear) and weigh that against the cost of collecting the toll (which includes the collection but also time cost)--
GRAPH: The "Optimal" Congestion Toll on a Roadway (just looking at time costs):
AC = time cost perceived by an individual user
MC = time cost of an extra vehicle on the road to all users
At first - there is no congestion - so charge where D = AC (and there is no MC from additional vehicles). But then with congestion -- marginal cost exists and the AC starts to increase.
So at Dn - there's no congestion, so need to charge a toll. (Remember, this is when the government is providing the road and deciding how to deal with congestion).
At D1 - there's congestion, so charge a toll = to the difference between AC and MC.
What about Mass Transit?
There are "peak" times when the congestion is a problem -- so do "peak pricing" --
GRAPH:
At D(low), MC is low, charge MC.
At D(high), MC has increased - so increase the price to where D or MWTP = MC. This will also lower demand during peak times (vs. one low price).