522 Econ Lecture 24 (April 30, 2009)
Final exam: Sunday May 10, 12:25 p.m., in 5206 Social Science HW3 will hopefully be returned on Tuesday, and my answers will be given out at the end of class today The final is cumulative everything through the end of today (but not from next week) is fair game but there will be more weight on the second half of the class Next week: Chao extra office hours Monday "optional" lecture Tuesday with some interesting additional material o much of it: behavioral law and economics (that is, how peoples' actual behavior differs from the standard economic models, and what this means for law and economics) no lecture Thursday, instead extra office hours o my usual OH 1:30-3:30 Wed o also 1-2:30 Thurs (during usual lecture time) o and some on Friday (hours TBA) Today: what have we done this year?
-1-
Well, we started out by defining efficiency, and asking, should the law be efficient? We gave a few reasons why it should Also pointed out some of the problems with efficiency But agreed that we'd focus on efficiency for most of the class. (Friedman offers three more reasons why we should focus on efficiency: o it's something that most people care about o some aspects of the law seem to have been designed to achieve efficiency, so thinking about efficiency helps us to understand them better o looking at efficiency is "one of the things economists know how to do and when you have a hammer, everything looks like a nail")
The choice of efficiency as our measure makes things much simpler in a lot of ways. For example, it means that: 1. we only have to focus on final outcomes who gets to own what, how many car accidents there are, etc. and not on process o that is, we assume people only care about what they get, not why they get it no concerns about procedural fairness, etc. 2. we can ignore transfers o that is, suppose that at some point, if a certain thing happens, our legal system is going to make Alice pay Bob $100 o a transfer of money has no direct effect on efficiency o so we can ignore the "direct" effect, and look only at the impact this has on incentives, and therefore on Alice and Bob's actions So we've decided to look at the law through the lens of efficiency This means that the "goal" will always be to maximize the total surplus or total wealth available to everyone in the world
-2-
This brings us to the first big recurring theme from this class the idea of getting efficient outcomes by setting up the law to make people internalize their externalities figuring out the efficient outcome requires considering the total social benefit and total social cost of an activity o that is, for efficiency, we want an activity to take place when its total benefit (to everyone) outweighs its total cost but, we expect a rational self-interested decision-maker to consider only their private benefit and private cost when deciding how to act when private cost = social cost and private benefit = social benefit, people naturally make choices that are efficient when private cost < social cost, an activity is imposing a negative externality, and people will tend to do it too much o for example, tragedy of the commons private cost to me of grazing my cattle on public land is lower than the social cost, so I do it too much o or: in a world without liability, my speeding imposes a cost on you, because I'm more likely to hit you so private cost of speeding is lower than the social cost so I speed too much (more than would be efficient) in order to fix this, we can write a law where I bear the cost of hitting you o this makes my private cost = social cost o so now I speed the efficient amount we saw this idea explicitly in contract law o when a promisor breaches a contract, he imposes an externality on the promisee o by setting expectation damages = harm done to promisee by breach, we force the promisor to internalize this externality and choose efficiently we saw this idea again in tort law o strict liability works exactly this way leads injurer to internalize the harms he does, leading to efficient precaution and activity levels we didn't discuss it in these terms, but we even saw this a little bit with Coase o suppose you like to make noise at night, but I live next door and like quiet o if you have a legal right to make noise, you have no reason not to o but suppose noise brings you joy of $50, but costs me $100 o if this right is tradable, keeping it costs you $100 in opportunity cost, since instead you could sell me the right
-3-
So that's one recurring theme designing the law to get people to internalize the externalities they cause, in order to get to efficient outcomes. Another recurring theme although we've discussed it in slightly different terms is the distinction between protecting a right versus protecting an interest suppose I own something my right to that thing is protected if nobody is allowed to take it away from me without my permission o or rather, since we can't control other peoples' actions... o my right is protected if anyone who violates it gets punished severely, so we expect nobody will do that on the other hand, my interest in that thing is protected if anybody who takes it away from me, has to reimburse me for the value I expected to get from it (protecting a right is done by injunction; protecting an interest is done by damages)
property law tends to protect our rights to things we own trespassing on my land is a crime if someone is caught trespassing, they don't just have to pay for the flowers they trampled, they may go to jail and stealing my iPhone is a crime if someone is caught stealing it, they don't just have to pay me its value, they again will go to jail but there are exceptions o if someone needs to use my cabin in the woods in an emergency, my interests are protected they'll have to pay for the food they eat, and replace the window they broke to get in but my rights are not absolute o and similarly, if the government wants to take my land, they only have to protect my interest in it by paying me its value (eminent domain)
tort law, on the other hand, tends to protect our interests certainly, I can be thought of as owning my arm but someone might hit me with a car and breaks my arm, and only owe me the amount of the damage my interest in that arm the value I would get from that arm is protected, but my right to that arm is not
-4-
in fact, we can think of that as the key difference between criminal and tort law because they are voluntary, crimes are things we may want to deter entirely so we give people a right to not get stabbed, and protect that right with a severe punishment torts, on the other hand, are often accidental reducing them to 0 is impossible, and reducing them almost to 0 is impossibly expensive so our goal is to reduce them to the efficient level and we do this by pricing them getting injurers to internalize the cost of their accidents, so that they make efficient choices as it happens, this is the same as protecting peoples' interests since they will be reimbursed the amount of the damage done
contract law tends to protect interests when someone breaches a contract, we saw that it's generally efficient for them to owe expectation damages this can be thought of as protecting the promisee's interest in the promise the value they expected to get rather than their right to the promise on the other hand, when a seller breaches a contract to sell a unique object, the remedy is often specific performance the court forces them to go through with the sale this means the promisee's right to the promise is protected the seller is not allowed to back out and pay damages
so we typically face a choice between protecting a right or an interest or, protecting an entitlement via injunctions or damages
-5-
We already said: designing an efficient legal system can be thought of as trying to maximize total surplus But alternatively, we can think of it as trying to minimize costs think of starting out at the most efficient outcome, and then think of costs as any deviations away from that efficient point costs generally take two forms o there are direct costs of implementing a particular aspect of a legal system o and there are indirect costs that aspect has, by leading to imperfect incentives and therefore less-than-perfectly-efficient outcomes o (we called the latter "error costs') the more complicated and complete a system, the higher the direct costs, but the lower the error costs so there is often a tradeoff between a "simpler" and a "more complete" system this is another theme that we saw in several different settings in this class
We saw this with Demsetz, who considered private ownership rights to land among Native Americans The direct cost: "boundary costs" the direct costs of clarifying whose land is whose, punishing trespassers, and so on The benefit: improved incentives not to over-hunt, and therefore more efficient use of the land o (reduction in "error costs" of overuse) Efficiency says we should have private land ownership when the benefit is bigger than the cost o when reducing error costs outweighs increase in direct costs Demsetz points out that the fur trade increased the benefit above the cost, so this is when Native Americans developed ownership rights over land We saw a similar tradeoff with whaling rules "Fast fish/loose fish" was a very simple rule, and very cheap to implement there were few disputes "Iron holds the whale" leads to more disputes, but also more efficient incentives when whales are too fast/violent to hunt with harpoons tied to ship So "Iron holds the whale" has higher direct costs, but lower indirect costs (more efficient outcomes); which one is better depends on the circumstance, that is, how severe the indirect costs are o with right whales, indirect costs were low simple rule worked "well enough"
-6-
o with sperm whales, indirect costs were high the old rule wouldn't provide enough incentives to hunt sperm whales We saw the same tradeoff between cheaper, simpler rules and more expensive, more complex rules when we compared injunctive relief to damages in property law Injunctions force the concerned parties to bargain among themselves But damages are more costly for the court to calculate When transaction costs are low, bargaining is likely to succeed, so injunctions are cheaper When transaction costs are high, bargaining may fail, so error costs of injunctive relief would be high o so the benefit of a damages system may outweigh the additional costs
We saw the same tradeoff when we considered the fees charged to initiate a lawsuit Higher filing fees mean fewer lawsuits, which means lower direct costs But higher fees also mean more accidents go "unpunished," which weakens incentives for precaution And we saw the same tradeoff with crime Catching and punishing more criminals is usually costly But it reduces the social cost of crime by reducing the crime rate
-7-
early in the class, we introduced Coase the idea: when there are no transaction costs (no impediments to rational people bargaining with each other), all we have to do is define property rights and make them tradable, then sit back and let people trade them until we get to efficiency but when there are transaction costs, this won't work so we face two options: we can try to reduce transaction costs or we can try to get the allocation right to begin with there are a number of possible sources of transaction costs: search costs difficulty in even finding a trading partner private information not knowing what something is worth to the other person large numbers of parties freerider or holdout problem makes it hard to reach agreement with a large group hostility enforcement costs
Property and nuisance law exist first to clarify peoples' rights/entitlements o so that they can enter into voluntary exchanges that are mutually beneficial. o different rules for allocating entitlements will differ according to how they handle situations where transaction costs are too high for efficiency to be reached "automatically." Contract law can be thought of as dealing mostly with situations where the costs to ex-ante negotiations are low, but the costs of renegotiating contracts under changed circumstances are higher o that is, we agree to a contract up front, but we don't want to have to renegotiate under every possible contingency Tort law can be thought of as dealing mostly with accidental situations, often between strangers, where ex-ante negotiations are impossible o or, the transaction costs of ex-ante negotiations are too high And criminal law can also be thought of as dealing with situations where ex-ante negotiations are impossible, but for other reasons
-8-
o in particular, many things that are crimes are thought of as being public bads so there are many parties being harmed, so transaction costs are high So that's the very high-level version. In a bit more detail... We began with PROPERTY LAW
We started out by motivating the need for property law: Without any notion of private property, publicly-owned resources will be overused (tragedy of the commons) And without enforcement of property rights, people will waste resources on "possessory acts": grabbing property (and protecting their own property from those who would grab it).
We addressed four key questions in property law: What can be owned? What can/can't owners do with their property? How are property rights established? What remedies are provided when property rights are violated?
and gave answers to each based on efficiency. What can be owned? General answer: private goods should be privately owned Tradeoff between efficiency gains from private ownership and costs of having a system in place to protect/enforce property rights. (Congestion costs versus the cost of barbed-wire fences.) Demsetz example of land rights among Native Americans appearing as the benefit from private ownership increased. Examples from intellectual property (patents, copyrights, trademarks, trade secrets), organizations versus corporations
-9-
What can/can't owners do with property? General answer: "maximum liberty" efficiency favors allowing owners to use their property in any way that doesn't impose an externality on others Even in situations with externalities that is, where one person's use of his property interferes with someone else's use of theirs who starts out with the right should affect final distribution of wealth, but not efficiency, when transaction costs are low But when transaction costs are high, who starts out with the right, and how it's protected, will matter for efficiency Two approaches: minimize transaction costs, so bargaining is more likely to succeed (Normative Coase), or allocate initial rights efficiently, so it doesn't have to (Normative Hobbes) We also examined some limitations of property rights, and economic rationale for each: o Inalienability. You have the right to vote, but can't give it to someone else; you own your kidneys, but can't sell them. o Can't always unbundle rights and transfer some parts without others o Can't limit future rights in perpetuity o Private necessity allows others to use your property in emergencies o Eminent domain allows government to take your property at "fair value" o Government is allowed to regulate your use of property, even when this reduces its value, at least up to a point.
How are property rights established? No general rule, but interesting examples from fugitive property whales, foxes, underground gas Often a tradeoff between having a simple rule (cheaper to enforce, might encourage bargaining) versus a rule which gives proper incentives but is more complicated or difficult to enforce. o Pierson v Post majority chose to enforce a bright-line rule (fox isn't yours till you catch it), but dissent argued hunting foxes is a good thing, law should create incentives to hunt o Whales: simple rule (Fast Fish/Loose Fish) adequate for certain types of whales, more subtle rule (Iron Holds the Whale) necessary for other types Another place this tradeoff arises: First possession versus tied ownership. Linking ownership to first possession leads to overinvestment in "possessory acts" (homesteading too early, etc.); tying ownership to ownership of other property is sometimes more complicated, and therefore more costly (or impossible, as with underground natural gas) Proving ownership Two ways to give up rights to something: adverse possession, estray
- 10 -
What remedy when property rights violated? Injunctive relief (property treatment) versus damages (liability) Calabresi and Melamed: property treatment more efficient when transaction costs are low, since it encourages voluntary exchange; damages are better when transaction costs are high, since it leads to the efficient outcome when voluntary exchange is impossible Temporary damages (for harm already done), permanent damages (to cover all future harm) Rules for injunctions versus damages, and temporary versus permanent damages, based on what is easiest for the court to measure accurately: o Private nuisances favor injunctions, since transaction costs tend to be low o Public nuisances favor damages, since transaction costs tend to be high o If damages are easy to measure and innovation occurs rapidly, temporary damages are more efficient o If damages are difficult (or costly) to measure and innovation occurs slowly, permanent damages are more efficient
- 11 -
Next, we came to CONTRACT LAW We motivated contract law as a way to enforce promises, to encourage cooperation in situations where transactions could not occur all at once o that is, where at least side one had to rely on a promise by the other. We described an early legal theory of contracts, the bargain theory o under the bargain theory, promises were enforced if they were given as part of a bargain o this depended on three elements: offer, acceptance, and consideration (the promisee giving something up in exchange for the promise) We came up with a general principle: efficiency depends on promises being legally enforceable if both parties wanted them to be at the time the promise was made
We then came up with a long list of ways that contract law can promote efficiency: 1. can enable cooperation (by turning games with noncooperative solutions into games with cooperate solutions) 2. can encourage efficient disclosure of information a. if I can promise to fix anything wrong with my car in the three months after you buy it, we can get past the private-information problem b. Hadley v Baxendale withholding the information about how time-critical the shipment was inefficient 3. can secure breach of contract only when it's efficient a. this works by making the promisor internalize the harm that breach causes the promisee, which is done using expectation damages 4. can secure optimal reliance a. that is, to get the promisee to invest, but not overinvest, in increasing the value of the promise 5. can decrease transaction costs of negotiating by providing efficient default rules a. usually by supplying defaults that most people would have agreed to had they considered the question ex-ante b. Ayres and Gertner: deliberately inefficient default rules (penalty defaults) may sometimes be valuable, since they can encourage information disclosure or force the parties to explicitly negotiate new terms c. (if the default rule goes against the better-informed party, they may provide information in order to negotiate a different default rule. Hadley, also real estate agents on who keeps earnest money when a sale falls through) 6. (5b) also, efficient regulations (rules that can't be overruled in the contract.) 7. to encourage enduring relationships/repeated interactions, which rely less on courts to sustain cooperation
- 12 -
We looked at a number of different remedies for breach of contract: expectation damages opportunity cost damages reliance damages specific performance We considered the incentives that each remedy created for breach and for reliance o and the effect of each one on each side's threat point if the parties were to renegotiate the contract. We discussed the question of how reliance investments were treated in damages o And the "paradox of compensation" that it is very difficult to set all the incentives correctly at the same time We discussed the fact that damages typically include "foreseeable reliance," which might be a reasonable proxy for "efficient reliance." We discussed penalty clauses in contracts, and the fact that they are often not upheld in court.
We looked at a number of formation defenses and performance excuses: incompetence a minor or someone crazy can't agree to a binding contract; contracts are enforced only when they are in these parties' best interest, to give the other side an incentive to look out for them. o but, being drunk (up to a point) does not get you out of a contract duress and necessity contracts are not enforced if they were signed under dire constraints o we saw the example of the sailboat in a storm, as well as the accident victim on the midterm the price that leads to the best incentive ex-ante is not necessarily the price that would be negotiated during the storm, so contracts signed then might be overturned unconscionability (terms that shock the conscience of the court) derogation of public policy, binding your hands in negotiations with another party defenses that the contract was predicated on bad information: o Fraud o Failure to disclose o Frustration of purpose o Mutual mistake
- 13 -
In each case, we looked for an economic justification for the doctrine, and generally were able to find them. We discussed efficient default rules an efficient default rule generally allocates a risk to the low-cost avoider o that is, a risk should generally be allocated to whichever party can bear the risk most cheaply, possibly by taking actions to reduce the risk we also introduced the principle of default rules which unite knowledge and control, which is generally efficient: o Contracts based on unilateral mistake are generally upheld such as someone not knowing what their antique car was worth when they sold it since these unite knowledge with control the party who knows more about antique cars now has the car, and would likely take better care of it. o On the flip side, Obde v Schlemeyer: the seller knew his building had termites; the usual view that he had no duty to disclose was overturned, since in this case the sale separated knowledge from control.
TORT LAW Next, we turned to tort law. The traditional theory of tort law has three elements: harm causation breach of duty (required under a negligence rule but not a strict liability rule) We introduced a simple model of unilateral precaution, and found that no liability lead the injurer to take inefficiently little precaution strict liability lead the victim to take inefficiently little precaution negligence rules (with appropriate standards of care) lead to both sides taking efficient levels of precaution We looked at several different negligence rules: simple negligence, under which a negligent injurer is liable negligence with a defense of contributory negligence, under which a negligent injurer is only liable if the victim was not negligent as well comparative negligence, under which the cost of the accident is split according to the relative size of each party's negligence In addition to precaution, we considered the effect of each of these rules on the activity levels chosen by the injurer and the victim.
- 14 -
(Instead of distinguishing between precaution and activity, Friedman distinguishes between observable precaution and unobservable precaution, which has the same effect.) We discussed the Hand Rule for determining what level of care constitutes negligence Hand held that a party was negligent if a higher level of precaution would have been cost-justified, that is, efficient This is determined by comparing the reduction in expected harm due to accidents, with the cost of the precaution For instance, a particular type of precaution is cost-justified if ( Pwith Pwithout ) X Harm per Accident > Cost Or, with precaution being a continuous variable, if - p'(x) X A > w
We looked at the effects of court errors on incentives: Under strict liability, random errors have no effect, but systematic errors (either in failing to hold some injurers liable, or in awarding incorrect levels of damages) lead to a different level of precaution Under a negligence rule, small errors in judgments have no effect on precaution, but systematic or random errors in applying the legal standard of care do have an effect
We also discussed tort law in a setting with sellers and consumers in a competitive market: Under a negligence rule, the seller will always take appropriate precaution, but if customers underestimate the risk, they will set an inefficiently high activity level Under strict liability, the seller will take appropriate precaution, and since the remaining risk will be built into the price of the good, customers will set the efficient activity level regardless of whether they perceive the risk correctly Under no liability... o If customers perceive each seller's risk level accurately, we will have efficient precaution and efficient activity o If customers perceive industry-average risk, but not each seller's risk, we will have too little precaution but efficient activity given the risk o If customers are oblivious to risk, we will have too little precaution and too much activity
- 15 -
We talked about relaxing the assumptions of the "standard" model, by considering... irrationality o for example, biases in how people perceive probabilistic events, and negligence caused by unintended lapses judgment-proof defendants, that is, defendants whose liability is limited by their limited assets and the ability to declare bankruptcy regulation, which punishes violations before an accident occurs, and can therefore be useful when defendants are judgment-proof insurance litigation costs, which can distort incentives for precaution if some victims won't find it worthwhile to sue We also considered some additional complications to our basic model: vicarious liability, in particular, the liability of an employer for his employees' actions joint and several liability when there are multiple parties responsible for the harm evidentiary uncertainty, which has similar effects to random errors in damages
We talked about the difficulty in assessing perfect compensatory damages in cases of injury or death, and some of the attempts to quantify the value people place on a life from decisions people make in other areas. We talked about punitive damages, which are generally only allowed when the injurer's behavior is "malicious, oppressive, gross, willful and wanton, or fraudulent." From a theoretical point of view, punitive damages can be used when successful prosecution is uncertain, in order to make expected damages equal to actual harm, to create correct incentives for injurer precaution.
We talked about the McDonald's coffee burn case, and a few empirical facts about the tort system in the U.S.
- 16 -
LEGAL PROCESS Next, we moved to examining the legal process itself We began with the assertion that an efficient legal process is one that minimizes the sum of two types of costs: o direct (administrative) costs o error costs, that is, the social cost of any distortions caused by judgments differing from the theoretically perfect ones We pointed out that in addition to the number and size of judgment errors, error costs depend on whether errors affect production or only distribution o that is, whether expected damages have a strong incentive effect or not. We pointed out that when plaintiff and defendant have similar estimates of the likely outcome of a trial, or are relatively pessimistic, out-of-court settlement seems likely But when the two sides are relatively optimistic, settlement may be impossible We discussed pre-trial exchange of information voluntary exchange, as well as mandatory exchange during the discovery process o We saw that voluntary information exchange should always promote settlement, while forced information may or may not We discussed nuisance suits o baseless lawsuits meant only to attract a settlement o use...