Back to the Table of
- I. Three
- A. The effect of legal rules.
- B. Explaining legal rules.
- C. Choosing rules
- II. Ideas:
- A. Efficiency.
- 1. Simply thought of as size of the pie, but ...
- 2. that is not an adequate explanation when the pie we
are measuring consists of everything that happens to
- 3. The question "what is the efficient rule and is it
the same as the actual rule" then ends up as the center of
- B. Why things are inefficient:
Externalities and related problems.
- C. And what you can do about it:
- 1. Direct regulation
- 2. Pigouvian tax
- 3. Property rights, transactions and transaction costs
- D. Relates to our three projects
- 1. The tendency of markets to produce efficient outcomes
is important to understanding what will happen--for example,
that contract terms will be efficient wrt the parties, so
restrictions tend to make the parties worse off.
- 2. Efficiency is a conjecture for explaining common law
rules, and ...
- 3. A candidate for recommending rules.
- III. Coaseian
lessons: A way of thinking about these issues:
- A. What we own are not things but
- 1. We can transact over rights, so ...
- 2. The question is not merely who ought to end up with
the right to do what, but ...
- 3. Who should start with it, given that we don't know
who should end up with it but do know a good deal about how
rights can move from one person to another.
- 4. Which depends on transaction costs
- B. There is no reason to expect that any of
the solutions will give us a fully efficient outcome.
- 1. In particular, the Pigouvian won't, in general.
- 2. Nor will the regulatory
- 3. Nor the Coaseian
- IV. We should be
thinking in terms of three mechanisms for dealing with
- A: Regulation
- 1. EPA regulations, etc., but also ...
- 2. Negligence rules.
- 3. More generally, any mechanism which depends on the
regulator knowing how the regulatee ought to behave (in our
context, what behavior is efficient) and making it costly
for him not to behave that way.
- B. Pigouvian:
- 1. damages
- 2. effluent fees, traffic tickets
- 3. More generally, any mechanism that forces one of the
people causing some costly outcome to bear the cost, giving
him an incentive to adjust his behavior accordingly.
- 4. Where damages also reduce the incentive of the
victim, who is usually also a cause.
- 5. Fines don't reduce it, but don't give the proper
- 6. Both parties are fined (examples?)
- C. Coaseian (property rights,
- 1. Typically associated with a property right, since the
right is then moved to its highest valued use by
transactions, but ...
- 2. One could use a property right to impose a regulatory
solution--if transaction costs were so high that whoever
started with the right ended with it.
- 3. And one can transact over liability rights--I might
agree not to sue you in exchange for a payment, for example.
- 4. Generally, any mechanism where we move towards the
efficient outcome via bargaining among the affected parties.
- 5. The norm for most of the economy--ordinary
property--and one option for externalities, etc.
- 6. Note that it includes cases where we bargain once to
get a permanent solution--for instance, the factory buys the
- V. Insurance issues:
allocation of risk runs through much of the law.
- A. Risk spreading.
- B. Moral hazard--or more generally, the
affect of risk allocation on the incentive to take efficient
precautions to prevent the undesirable outcome.
- C. Adverse selection--more generally, the
effect of risk allocation if a transaction occurs on the
incentive to make efficient transactions.
- D. These desiderata may pull in opposite
- E. In which case you either settle for the
least bad solution, or ...
- F. Bring in additional mechanisms. For
- 1. Reputation plus caveat emptor solves moral hazard on
both sides--if consumers are well informed.
- 2. Liability plus insurance combines optimal moral
hazard incentives with optimal risk spreading in the case of
liability for death or crippling injury.
- G. Note that we have a choice in dealing
with such problems among the three approaches described
- 1. Regulatory approach--producers will take the
following precautions (OSHA).
- 2. Pigouvian approach: The legal rule is caveat emptor
(or caveat venditor, or some mix).
- 3. Coaseian approach: Freedom of contract. Waivers and
guarantees both enforced.
- H. And note that which makes sense depends
largely on transaction costs:
- 1. In the product liability context, they are low, so
there is a strong argument for freedom of contract.
- 2. But in the tort context, transactions are usually so
costly that they don't occur, so we are stuck with the
- VI. Ex ante/Ex Post
another issue that runs through the law.
- A. Simple question: Speed limits vs
punishment for being in an accident.
- 1. Ex post better utilizes the actor's private
information--I know I'm not paying attention but the cop
- 2. Ex ante permits the use of smaller punishments
applied more often.
- 3. So you always use some ex post, but when the size of
ex post gets big enough so that punishment costs start to
shoot up, you may start adding in some ex ante.
- 4. Which is how we handle traffic accidents.
- B. The same analysis explains the paradox
of redundant causation: Two hunters independently and
accidentally shoot a third.
- 1. On the fact of it, neither should be liable, since
neither did marginal damage.
- 2. But that is not what actually happens, nor should it
- 3. Because we know that we underpunish the tort of
accidentally shooting people, since adequate punishment
would be (impossibly?) costly, and ...
- 4. Punishing people in the two bullet case imposes an
additional incentive not to take actions that might lead to
- 5. In other words, the two bullet outcome is a signal of
behavior that you want to punish ex ante, and you do so by
punishing the (harmless) two bullet case ex post.
- C. And explains why we punish unsuccessful
- 1. Again, in a world of costless punishment we are
better off only punishing successful attempts.
- 2. But in a world of costly punishment we add an ex ante
punishment for trying to kill you, measured by the
unsuccessful attempt, to the ex post punishment for
- 3. And, in fact, "attempts" are criminal but are not
- 4. Which fits the fact that tort law uses more efficient
punishements than criminal law and (more important)
punishments whose inefficiency does not rise with their size
(i.e. litigation cost as a fraction of amount at stake
probably does not increase with amount at stake).
- A. Why are (and should be) some things
property, some commons?
- 1. Because of costs associated with defining, defending,
transacting over, obtaining
- 2. Balanced against cost of inefficient production,
overuse, etc. if things are not property
- B. One application is in understanding
property rights in primitive societies.
- C. Another is to intellectual property. In
- 1. Why have long protection given easily for copyrights.
- 2. Short protection with difficulty for patents.
- 3. Treat computer programs as writings.
- A. Initial puzzles:
- 1. Why have contracts? Because transactions occur over
time, and the parties are locked in once the process starts.
- 2. Why have state enforced contracts? Because although
other mechanisms exist and are often functional, even
superior (diamond market in NY), they don't cover all of the
- 3. Why have contract law--instead of simply enforcing
all contracts as written?
- a. Because we need some way of knowing if a contract
exists. One sided, no consideration, mistake, fraud,
- b. And of filling in the blanks,
- c. And we may not want to enforce all contracts as
written. Economic argument in favor of doing so--but
exceptions for third party effects, infants, senile, ...
- B. Making sense of duress
- 1. Real duress. Enforceability helps you when you need
to pay off the mugger, but makes mugging more likely.
Whether it is desirable depends on the relative size of the
two effects. Ransome in war. Surrender mechanisms in
- 2. Sinking ship case. Not real duress, but bilateral
bargaining with high costs and no clearly efficient rule.
- 3. Form contracts.
- a. Even without bargaining, incentive to write them
- b. Provided that the other party either reads them or
knows your reputation.
- c. And there are lots of good, innocent reasons to
use form contracts.
- C. Filling in the blanks.
- 1. Parties will tend to bargain to the efficient (for
them) outcome, so ...
- 2. Fill in with efficient rules if you want to produce
what they would have agreed to, or ...
- 3. If you don't want them to have an incentive to write
longer contracts next time.
- 4. But one can think of reasons why some judges might
reject that argument.
- D. Our analysis of insurance helps tell us
how to draft and fill in the blanks where the issue is
allocation of risk.
- E. What about the result of breach?
- 1. No damages or specific performance throws us back on
Coaseian bargaining to get the efficient outcome.
- 2. Expectation or reliance damaages are Pigouvian
solutions, each of which results in some inefficiencies for
reasons implied by Coase's critique of Pigou.
- 3. Liquidated damages pushes the Coaseian bargaining to
before there was a contract, when there is less of a
bilateral monopoly problem--but at the cost of requiring you
to set the damages without the information provided by what
happens between contract formation and breach.
- 4. Penalty clause is essentially a contractual property
rule--like specific performance, but weaker.
- F. Marriage contract:
- 1. Exists because of sunk costs, relationship specific
investments (esp children).
- a. Presumably became less important as those
decreased, but ...
- b. Still a lot of puzzles and alternative theories as
to why the drastic changes of the past fifty years
- 2. Economics of wedding rings.
- 3. Problems of designing the contract.
- a. The usual problems with all the breach
- b. Made harder by the fact that performance is hard
- c. And the contracts are made by inexperienced
- G. Adoption, regulation of sex, etc.
- 1. Interesting commodification argument, but ...
- 2. Seems to fly in the face of first amendment
- 3. Do children impose net externalities? Not clear.
- IX. Tort
- A. What is is: Private action for a
- B. Four issues
- 1. What makes it wrongfultort involves a violation.
- 2. Causation: What does it mean to say A caused the
- 3. Liability--strict liability, negligence, contributory
- 4. Damages--how calculated?
- C. Wrongful ultimately means "undesirable
of a sort best handled by tort law!"
- D. Causation ultimately means: Causally
related in a way such that making the causer liable has good
consequences. Hence not in the coincidental causation case, yet
(paradoxically) in the redundant case. Many other puzzles such
as probabilistic cases.
- E. Liability: Different rules involve
different ways of solving the dual causation problem by
combinations of regulatory and Pigouvian approaches. Coasian
approach largely out because of high transaction costs for many
(but not all) tort situations. Objective--optimal precautions,
supply of accidents etc.
- F. Damages. Simple rule--damage
- 1. Analysis becomes more complicated with undertainty
conviction, costly punishment and apprehension.
- 2. Which may, or may not, explain punitive damages. Lots
of possible explanations, none of which fit perfectly what
- G. Damages for loss of life--read my
- X. Crime:
- A. Like tort, but enforced by the
- B. Raises issue of optimal enforcement,
punishment, given that both are costly.
- C. And the further complications if we
allow for the incentives of the enforcers--which may be an
argument against efficient punishments.
- D. Gets us into the question of why have
- 1. Best answer: Judgement proof defendants, negative
- 2. Solved by making deterrence a private good.
- 3. But not for anonymous victim offenses.
- E. Ethics:
- 1. We don't like criminals.
- 2. Should their gains count? Yes because ...
- 3. But not liking them is rational--piling moral
sanctions on because other punishments are costly, hence
lower than we would like.
- XI. Antitrust
- XII. Alternative
- A. What we are used to is not the only
- 1. Iceland at one extreme
- 2. 1th c. England
- 3. Shasta County
- B. All teach us lessons about the options
and their problems.
- XIII. Is the common
law efficient? The draft of my Posner article is available to any
of you who are interested.
- A. A very productive conjecture.
- B. No good theoretical arguments for
- C. Some but very fuzzy empirical evidence
Back to Part I of the
Back to Part II of the
Back to the class home page