Today’s post is from chapter 15 of the textbook and is on economic property rights and transaction costs. So far one of our assumptions for our model has been that markets work for free. There is obviously a problem with this and so this post will go into the idea of transaction costs and what they mean for the economic model.
First we will go a little bit away from this and talk about the Coase theorem. The Coase theorem is that the allocation of resources is independent of distribution of property rights when transaction costs are zero. To illustrate this the textbook uses the example of a rancher and a farmer.
As in the textbook “Consider two worlds, both made up of farmers, ranchers, and no fences. Everything in the two worlds is exactly the same – except one thing. In one world the rancher is liable for the damage his cattle cause as they trespass on the farmer’s field. In the other world he is not liable. Both worlds are characterized by zero transaction costs; that is, it costs nothing for the farmer and rancher to trade with one another. Would the number of cattle be different in either world?”
The answer by Ronald Coase in 1960 was no it would not be different. To show this we will continue on with the example from the textbook. Consider the following table.
|
Number of Steers |
Marginal Private Cost |
Marginal Crop Damage |
Marginal Social Cost |
|
1 |
$100 |
$100 |
$200 |
|
2 |
$150 |
$100 |
$250 |
|
3 |
$200 |
$100 |
$300 |
|
4 |
$250 |
$100 |
$350 |
|
5 |
$300 |
$100 |
$400 |
|
6 |
$350 |
$100 |
$450 |
“The figures in column 2 indicate the “private” cost to the rancher of producing steers. These would include, for example, feed, cowboys, etc., meaning all the costs to the rancher of raising cattle to maturity and bringing them to market. According to the table, it costs the rancher $100 to raise one steer, an additional $150 to raise a second, $200 more to raise a third, etc. Not included in these costs, however, are the damages imposed on the neighboring farmer. Each steer tramples $100 worth of crops during its lifetime. These costs, which are “external” to the rancher, are often referred to as “externalities”. Thus, the actual cost to society of raising 1 steer is not just the $100 diverted from other uses by the rancher, but also the $100 in crops that never get harvested, yielding a true marginal cost of $250. The complete, actual cost to society of producing a good is generally referred to as its “social cost”. Likewise, “marginal social cost” refers to the true, complete cost of producing an additional increment of the good. The “private” cost is really a misspecification of cost; it leaves out part of the true effects of producing this good: in particular, the costs imposed on someone else. We consider private costs because it often seems that producers do not in fact always bear all of the costs of their decisions.
Suppose now the market price of mature steers is $300. How many steers will the rancher produce, and how does the answer depend, if at all, on the rancher’s liability for crop damage caused by his steers? Assume first that the rancher is liable for all such crop damage. In this case, the figures in column 4, marginal social cost, are also the rancher’s own marginal costs of production. For each steer produced, the rancher must compensate the farmer $100 to cover the cost of the ruined crops. The rancher produces 3 steers under these constraints. Production is carried out until marginal cost rises to the market price. The rancher makes $100 on the first steer, $50 on the second, and zero on the third, but, as usual, we assume production is carried out to this limit. He receives total rents of $150 on this production
Suppose now the rancher is not liable for crop damage. In this case, the rancher does not have to compensate the farmer for the $100 of crops each steer destroys. It appears the “private” marginal cost figures in column 2 will determine output. In that case, the rancher would produce 5 steers, where the private marginal cost equals the market price. Coase showed, however, that this conclusion depends on the assumption that the farmer and rancher are unable to negotiate a mutually beneficial contract with each other.
When the rancher produces a fourth steer, his rents are potentially $50: the difference between the market price, $300, and the private marginal cost of that steer, $250. However, this steer produces a greater amount of damages, $100, than the rents received. The farmer would benefit by $100 if the steer were not produced; the rancher gains only $50 from producing it. In this situation, where the loser loses more than the gainer gains, the loser can pay the gainer something greater than the potential gain and less than the potential loss, and the position of both parties will be improved. For example, if the rancher accepts a payment from the farmer of $75 to not produce the fourth steer, then the farmer and the rancher each gain $25. Similarly, since the rancher makes no rents on the sixth steer, any small payment from the farmer will induce him not to produce it, thereby saving the farmer almost $100 in the process. If damages are being produced by someone’s actions, those costs must be weighed against the benefits of the actions. If the costs are larger than the benefits, the parties can contract with each other to avoid these losses. Assuming, therefore, that the cost of transactions is sufficiently low that the farmer and the rancher can negotiate, resource allocation is the same, regardless of the assignment of liability. In this example, the rancher produces 3 steers under either assignment of liability. The wealth of the farmer and rancher are of course affected by who has to pay whom. If the rancher is liable for crop damage, he is worse off and the farmer is better off than if the rancher were not liable. The production outcome is the same, however, in either case: 3 steers.”
So this is what happens in a world with no transaction costs. Before we get into the meat of talking about transaction costs first we need to talk about economic property rights. Economic property rights is one’s ability to freely exercise a choice. These differ from legal property rights which is one’s ability under the law to freely exercise a choice and natural property rights which are one’s ability under nature to freely exercise a choice.
A property right is considered complete if you are able to make all of the decisions with respect to a good. If is perfect, if on the dimension you are choosing, there is no infringement on the choice you are making. Quite often our ability to make choices is limited, and in these cases we would say our property rights are limited. If our choices are totally limited, then we have no property rights at all.
Sometimes our economic property rights are incomplete because someone else is the true holder of the rights, such as with a passport. Other times our rights are imperfect because it does not pay to enforce them. When rights are too costly for anyone to own, the asset in question is said to be in the public domain. This is just the state where property rights no not exist. Transaction costs, what this chapter is about is just the cost of establishing and maintaining property rights.
Earlier we noted that individuals maximize utility and firms maximize profit. None of this changes with transaction costs. However there is a slight modification. Both are not net of transaction costs. This leads to the last economic principle in the book, optimal organization.
Optimal organization is that all economic organizations are designed to maximize the gains from trade net of transaction costs. So what causes transaction costs?
The underlying theme in understanding transaction costs is the notion of ignorance. Negotiation, fraud, communication and contracts all come about because knowledge is incomplete and not common. However information costs are not the same as transaction costs.
For transaction costs to exist, it must be costly to acquire information about anything: goods, people, and institutions. This means that information costs are a necessary condition for transaction costs to exist.
It is necessary however, to do more than assume costly information in order to generate transaction costs, because costly information merely makes for risky events. An additional assumption is required that enhances the problem of costly information. This is the idea that goods are not simple but that they are both variable and alterable.
The distinction between variability and alterability can be thought of as changes brought about by nature and by man. When goods are both variable and alterable then cheating becomes possible and this is where transaction costs arise from.
There are three different types of property and the transaction costs for each type vary. There is private property, common property and open access. Private property is what most of us think of and when the owners of private property decided how an item is to be used, who is to use it, and what happens to the income or utility of the item. Common property is where access to a good or resource is limited to some group, but within the group no one has the right to exclude others. Finally open access is a situation where no one has the right or ability to exclude anyone.
Each one of these different property right regimes has different benefits in terms of the wealth they generate. However, each one has different transaction costs associated with them as well. Which one is chosen as the optimal form of ownership depends on which one maximizes the value of the asset, net of these transaction costs.
An example of this that is commonly used is that of a fishing boat. Suppose that there is a village next to a lake where people fish. If you stand on the shore and fish you will catch 4 fish over the course of the day. However there is also a boat in this village and the number of fish caught in the boat will depend on how many people are in it.
When the boat is privately owned, the owner of the boat decides how many workers will be allowed on. We can think of the boat owner as hiring workers for the boat. Since workers can catch 4 fish on shore, the boat owner must pay each fisherman 4 fish to come fish on the boat. How many fishermen will be hired? We know the answer to this question is determined by where the marginal product just equals the wage. Since the boat allows more fish to be caught there will be some surplus that will go to the owner of the boat and under private property this surplus will be maximized. Hence private property maximizes the gross value of a resource.
Suppose now that no single person owns the boat, but the boat is owned in common by a small group of fishermen. These fishermen share the catch, and decide how many fishermen should enter the boat. Under these circumstances, the owners of the boat don’t want to put men in the boat until the wage equals the marginal product. What they want to do is maximize the average product because that’s what each one of them gets. This will no longer be equal to when the marginal product equals the wage but will in fact be before that. Because the number of fishermen has been reduced, along with the total catch, there is a deadweight loss associated with the common property ownership of the boat. In short, common property does not maximize the value of a resource.
Finally there is the example of open access. In this case it is quite easy. With open access, no one owns the boat. Anyone who wants to jump in and go fishing can do so. When this property right structure exists, people will enter the boat as long as the average number of fish they catch on the boat is equal to the number they can catch on shore, in this case 4 fish. Open access drives the value of a resource to zero, in this case they are catching just as many fish with the boat as without.
If this was all there was to the concept then obviously private property would be the best solution. However when we introduce transaction costs the picture is a little bit different. For the next little bit we will ignore common property and just assume that assets can either be in the public domain or held as private property.
We may have two questions here, what happens to the optimal ownership of the asset as the value increases, and is there an optimal value for the asset to be? The first question was initially raised by Harold Demsetz who argued that private property rights are established when the benefits of establishment out weight the costs of establishment.
For linear transaction costs, this means that the cost to protect the good eventually is less than the value of the good and at a certain threshold the good becomes worthwhile to protect and so becomes private property.
For non-linear transaction costs, which is what is the norm in reality, the more valuable an asset is, the more it costs at the margin to protect it. This means that it still holds true that as an asset value increases it will become worthwhile to protect it and it will move from the public domain to private property. However it is now also the case that at the upper end of value, items will have such extreme transaction costs that they will no longer be worthwhile to protect and as such will revert from private property to public domain.
Wealth maximizers do not like assets in the public domain. When assets are in the public domain because they are too costly to protect there are two general solutions available. The first is to innovate on the transaction cost margin. The second solution is the lower the gross value of the asset. An example of this would be making something look less valuable in order to make it look less attractive to thieves.
This is the basic ideas behind private property and transaction costs.
econsummerstudy 