Econ 522
Economics of Law
Dan Quint
Fall 2012
Lecture 20
Discussion question – recapping tort law
punish the choice
punish the outcome
• criminal law
• strict liability rule
• regulations
Choice
+
Bad Luck

Outcome
punish the combination of choice and outcome
• negligence rule
QUESTION: What are the pros and cons of each approach?
1
Monday’s
Experiment
2
Experiment
You have been asked to serve on a jury on a lawsuit dealing with
personal injury. In the case before you, a 50-year-old construction
worker was injured on the job due to the negligence of his employer.
As a result, this man had his right leg amputated at the knee. Due to
this disability, he cannot return to the construction trade and has few
other skills with which he could pursue alternative employment.
The negligence of the employer has been firmly established, and
health insurance covered all of the related medical expenses.
Therefore, your job is to determine how to compensate this worker for
the loss of his livelihood and the reduction in his quality of life.
3
What were we trying to test?
Half of you were asked…
The other half were asked…
(a) Should the plaintiff in this case be
awarded more or less than
$10,000?
(a) Should the plaintiff in this case be
awarded more or less than
$10,000,000?
(b) How much should the plaintiff
receive? (Please give a number.)
(b) How much should the plaintiff
receive? (Please give a number.)
(c)
(c) Are you male or female?
Are you male or female?
 Testing how much did the “suggestion” in question (a)
affect the answers to question (b)?
4
Well? Did the suggestion matter?
100%
80%
60%
40%
20%
0%
0
1,000,000
2,000,000
3,000,000
4,000,000
5
Well? Did the suggestion matter?
100%
80%
60%
40%
20%
0%
0
1,000,000
2,000,000
3,000,000
4,000,000
average: 730,981
6
Well? Did the suggestion matter?
100%
80%
60%
40%
10k
10MM
20%
0%
0
5,000,000
10,000,000 15,000,000 20,000,000 25,000,000
730,981
7
Well? Did the suggestion matter?
100%
80%
(plus one at $50 MM)
60%
40%
10k
10MM
20%
0%
0
730,981
5,000,000
10,000,000 15,000,000 20,000,000 25,000,000
6,300,000
8
Well? Did the suggestion matter?
100%
80%
(plus one at $50 MM)
60%
40%
10k
10MM
20%
0%
0
730,981
5,000,000
10,000,000 15,000,000 20,000,000 25,000,000
6,300,000
9
Well? Did the suggestion matter?
The only statistical test one ever needs is the IOTT, or “interocular trauma test.” The result just hits one between the eyes.
- David H. Krantz
Asked 10,000
Asked 10,000,000
Minimum
20,000
100,000
Maximum
3,270,420
50,000,000
Average
730,981
6,300,000
Median
350,000
3,000,000
Geometric Mean
365,651
2,627,402
10
Another way to see it
35%
30%
25%
20%
asked 10k
asked 10MM
15%
10%
5%
0%
< 150k
150k 400k
400k - 1.5MM - 4MM - > 10MM
1.5MM
4MM
10MM
11
Another way to see it
35%
30%
25%
20%
asked 10k
asked 10MM
15%
10%
5%
0%
< 150k
150k 400k
400k - 1.5MM - 4MM - > 10MM
1.5MM
4MM
10MM
12
Yet another way to see it
60%
50%
40%
asked 10k
asked 10MM
30%
20%
10%
0%
< 400,000
400,000 4,000,000
> 4,000,000
13
What about men versus women?
100%
80%
(plus one at $50 MM)
60%
40%
10k, male
10k, female
10MM, male
10MM, female
20%
0%
0
5,000,000
10,000,000 15,000,000 20,000,000 25,000,000
869,565
5,461,111
392,222
7,810,000
14
What do we learn?
 Nobody knows exactly what a leg is worth




Even among people asked exactly the same question…
…answers varied from $20,000 to $3,000,000…
…or from $100,000 to $50,000,000
Supports the point from Monday: incentive to fight!
 Since no one knows the right answer, “suggestions” matter



Higher suggestion raised average from $731k to $6.3MM
Reference point bias
Framing effects
15
Thinking about
the legal process
16
Over the last three months, we have…
 Developed theories of property/nuisance law, contract law,
and tort law
 Looked at how rules of legal liability create incentives
 Thought about how these rules can be chosen to try to
achieve efficient outcomes
17
We’ve been thinking of normative goal of
minimizing social costs
 Property law


Goal was to allocate resources/entitlements efficiently…
…or, to minimize inefficiencies due to misallocation
 Contract law


Goal was to further facilitate trade…
…or, to decrease inefficiencies due to unrealized Kaldor-Hicks
improvements
 Tort law


Goal was explicitly to minimize social costs…
…which consist of cost of accidents plus cost of precaution
18
Implicitly, we’ve generally been assuming
two things so far
 The legal system works flawlessly


Whatever theoretical goal we set, we can implement correctly
(In tort law, we’ve considered effect of errors)
 The legal system costs nothing
 Gave us nice theoretical results for achieving efficiency


Example: injunctions when TC low, damages when TC high
Example: strict liability when injurer activity matters a lot;
negligence when both sides’ precaution matters a lot
 Next: what additional concerns are there when trying to
put a legal structure in place to enforce these ideas?
19
What is the goal of the legal system itself?
 Start with the best possible benchmark


Theoretically perfect rules, implemented flawlessly and costlessly
That’s obviously the best we can hope to do
 How does reality differ?





1. Rules actually implemented won’t be the perfect ones
Imperfect rules will lead to imperfect incentives, leading to
less-than-perfectly-efficient actions and outcomes
Think of any loss of efficiency due to imperfections in legal system
as error costs
2. Actual system won’t be costless – administrative costs
Goal of legal system: minimize sum of these two costs
20
Administrative costs and error costs
 Administrative costs


Hiring judges, building courthouse, paying jurors…
More complex process  higher cost
 Error costs





Any legal process is imperfect
Errors are any judgments that differ from theoretically perfect ones
An error in computing damages after the fact only affects
distribution, not efficiency
But anticipated errors affect incentives, which may lead to
actions which aren’t efficient
Error costs are costs of distortions in actions people take
(precaution, activity levels, etc.) due to flaws in legal system
21
The goal of the legal process
 So theoretically, the efficient legal process is the one that
minimizes the sum of…


The direct costs of administering the system, and
The economic effects of errors due to process not being perfect
 We’ve already seen the tradeoff between these two types
of costs


Tradeoff between “simpler” versus “more complex” rules
We’ve seen this several times
22
We’ve already seen tradeoff between
administrative and error costs
 Whaling law – “fast fish/loose fish” vs. “iron-holds-the-whale”


FF/LF: lower administrative costs (fewer disputes)
IHTW: lower error costs (better incentives for whaling)
 Pierson v. Post (fox hunt case)


Majority: first to catch, otherwise “fertile source of quarrels”
Dissent: first to chase, hunting foxes is “meritorious”
 Privatizing ownership of land




Expanding property rights adds admin costs (boundary maintenance)
But lowers error costs (better incentives for efficient use of resource)
Demsetz: privatize when gains outweigh costs
Same as: pick system with the lower sum of admin + error costs
23
The legal
process
24
The legal process
 Once an accident has happened…

Victim could sue or not sue

The victim and injurer might quickly settle out of court

If the case proceeds to trial, the first step (in the U.S.) is a pre-trial
exchange of information

After that, victim and injurer might still settle out of court

If the case goes to trial, victim (now plaintiff) might win or lose

Losing side at trial can choose to appeal (or not)
25
Sue or not sue?
 Worth it for victim to sue if
Expected
value of
legal claim
 Probability of winning
at trial, times expected
judgment…
 Or likelihood of a
settlement, times
expected amount…
 Minus costs expected
to be incurred
>
Cost to
initiate
lawsuit
 “Filing fees”
26
Filing fees
 Expected value of claims should vary widely
Probability
Filing Fee
SUE
DON’T
SUE
Expected value of claims
27
How high should filing fees be?
 Recall the efficient legal system minimizes the sum of
administrative costs and error costs

Higher filing fees  fewer lawsuits  lower administrative costs

But, higher filing fees  more injuries go “unpunished”
 greater distortion in incentives  higher error costs

Filing fee is set optimally when these balance on the margin:


Marginal cost of reducing fee = marginal benefit
Administrative cost of an additional lawsuit = error cost of providing no
remedy in the marginal case
28
How high should filing fees be?
 Error costs



If we’re only concerned with efficiency, we don’t care about
distributional effects
That is, we don’t care if a particular victim is or isn’t compensated
So the size of error costs depends on how much peoples’ behavior
responds to the incentives caused by liability
 “The social value of reducing errors depends on whether
the errors affect production or merely distribution”


When errors have large incentive effects, filing fees should be low
When errors have small incentive effects, efficiency requires higher
filing fees
29
Filing fees
Probability
Filing Fee
SUE
DON’T
SUE
Expected value of claims
 As long as there are any filing fees or other costs to
litigation, some harms will be too low to justify a lawsuit

When harm is small to each individual but large overall, one
solution is a class action lawsuit
30
Class Action Lawsuits
 One or more plaintiffs bring lawsuit on behalf of a large
group of people harmed in a similar way

Example: California lawsuit over $6 bounced-check fee
 Court must “certify” (approve) the class


Participating in a class-action suit eliminates victim’s right to sue on
his own later
If suit succeeds, court must then approve plaintiff’s proposal for
dividing up the award among members of the class
 Class-action suits are desirable when individual harms
are small but aggregate harms are large…


Especially when avoidance of liability has strong incentive effect
But there’s also a danger
31
Number of lawsuits
Number of
lawsuits
not worth suing
for most victims
more precaution 
fewer accidents
Typical level of damages
32
Costly
litigation
33
An example from Polinsky, “An Introduction
to Law and Economics”
 I hit you with my car and did $10,000 worth of damage





We both know I was negligent
But courts aren’t perfect
If we go to trial, 80% chance I’ll be found liable, 20% I won’t
If I’m held liable, damages are correctly set at $10,000
So on average, if we go to trial, you expect to recover $8,000
 But if we go to trial, we both have to hire lawyers



Suppose this costs us each $3,000
Now your expected gain from going to trial is $8,000 – 3,000 = 5,000
And my expected cost is $8,000 + 3,000 = 11,000
34
An example from Polinsky, “An Introduction
to Law and Economics”
 So…


Going to trial gains you $5,000 (in expectation)
Going to trial costs me $11,000 (in expectation)
 If all this is common knowledge, maybe we can avoid trial


If we settle out of court and I pay you any settlement between
$5,000 and $11,000, we’re both better off
So maybe this happens
 But…
35
An example from Polinsky, “An Introduction
to Law and Economics”
 Suppose I’m more optimistic about my chances than you





You think I’m 80% likely to be found liable
I think I’m only 10% likely to be found liable
You think your expected gain is $8,000 – 3,000 = $5,000
I think my expected cost is $1,000 + 3,000 = $4,000
Now an out-of-court settlement is impossible
 On the other hand, if I’m more pessimistic than you,
settlement is even easier

But even then, not guaranteed if threat points are private information!
36
Failures in negotiations due to private
information – example
 Defendant made faulty product, lots of people injured





Some sustained minor injuries ($2,000)
Some sustained major injuries ($10,000)
Before trial, defendant can’t tell scope of plaintiff’s injuries
Suppose legal costs are $500 for each side
Suppose half of plaintiffs had major injuries, so average injury = $6,000
 Suppose defendant makes settlement offer of $6,000 to all victims
The ones with minor injuries will take …
 …and the ones with major injuries will go to trial and win $10,000

 Defendant has two choices:

Offer settlements large enough that everyone will accept
 But then even people with very minor injuries, or none, might sue
 Or offer only small settlements, and get stuck going to trial in many cases
37
Why does costly litigation matter?
 Under strict liability…




We said injurers internalize cost of accidents  efficient precaution
But this assumes cost of being sued = damage done
If courts are unpredictable and litigation is costly, private cost of
being sued for damages could be > or < cost of accident
Which could lead to too much or too little precaution
 But also…



If settlement talks break down and cases go to trial…
…then social cost of an accident includes both the harm done,
and the resources expended during the trial!
If trial costs $6,000, then social cost of the accident isn’t $10,000,
but $16,000 – which increases the efficient level of precaution
38
In the U.S., before going to trial, the two
parties must exchange information
 Disclosure (“discovery”) rule in the U.S. very extensive



Parties reveal basic arguments they’ll make, evidence that supports
them, names of witnesses, nature of each witness’s testimony
Each side can inspect other’s evidence, question its witnesses
Witnesses or evidence not disclosed during discovery may not be
allowed at trial
 Most European countries have little or no pre-trial discovery



Europe: juries rarely used in civil cases
Delays and interruptions less costly, more common
Under civil law, judges take more active role in developing arguments
and exploring evidence
39
Exchange of information might facilitate
out-of-court settlement
 Cooter and Ulen:
Trials occur when the parties are relatively optimistic about their
outcome, so that each side prefers a trial rather than settlement on
terms acceptable to the other side.
When the parties are relatively optimistic, at least one of them is
uninformed.
Pooling of information before trial that reduces relative optimism
promotes settlement.
Furthermore, by revealing private information to correct the other
side’s false optimism, the party making the disclosure increases the
probability of settling on more favorable terms.
40
But some exchange of information would
happen voluntarily anyway
 Involuntary disclosure leads to information being shared
that the parties would otherwise choose to withhold


This is usually information that corrects relative pessimism
So forced disclosure may make settlement less likely
 On the other hand, involuntary disclosure reduces
uncertainty, makes two sides’ threat points more clear

May make reaching a settlement more likely
 So overall effect is unclear
41
With costly litigation comes possibility of
“nuisance suit”
 Nuisance suit – lawsuit with no legal merit, purely meant to
extract an out-of-court settlement





Suppose trial costs $10,000 for plaintiff but $50,000 for defendant
If case goes to trial, plaintiff will get nothing
Threat points are -10,000 and -50,000
Gains from cooperation if settlement reached are 60,000
If gains are split evenly, defendant pays settlement of $20,000,
even though case had no merit
42
Who pays the costs of a trial?
 In U.K., loser in a lawsuit often pays legal expenses of
winner


Discourages nuisance suits
But also discourages suits where there was actual harm that may
be hard to prove
 In U.S., each side generally pays own legal costs

But some states have rules that change this under certain
circumstances
43
Who pays the costs of a trial?
 Rule 68 of Federal Rules of Civil Procedure
“At any time more than 10 days before the trial begins, a party
defending against a claim may serve upon the adverse party an
offer [for a settlement]…
If the judgment finally obtained by the offeree is not more
favorable than the offer, the offeree must pay the costs
incurred after the making of the offer.”
 “Fee shifting rule”
 Example




I hit you with my car, you sue
Before trial, I offer to settle for $6,000, you refuse
If you win at trial, but judgment is less than $6,000…
…then under Rule 68, you would have to pay me for all my legal
expenses after I made the offer
44
Who pays the costs of a trial?
 Rule 68 does two things to encourage settlements:


Gives me added incentive to make a serious settlement offer
Gives you added incentive to accept my offer
 But not actually as generous as it sounds

Attorney’s fees not always included in fees that are covered
 Asymmetric


Plaintiff is penalized for rejecting defendant’s offer
Defendant is not penalized for rejecting offer from plaintiff
45
Trial
 In Europe…


Judges in civil trials take active role in asking questions and
developing case
“Inquisitorial system,” since judge asks questions
 In U.S…



Lawyers’ job to develop case
Judge is more of a passive referee
“Adversarial system,” since competing lawyers are adversaries
46
Unitary versus Segmented Trials
 Trial has to answer two questions:


Is defendant liable?
If so, how much are damages?
 Unitary trial considers liability and damages at same time

Economies of scope
 Segmented trial considers liability first, then damages
later (if necessary)

Damages phase may not be necessary
 In U.S., judges have discretion over which type of trial
47
Rules of evidence
 Rules for what evidence court can pay attention to
 Textbook gives examples where rules seem inconsistent, if
goal is simply to maximize probability of “right outcome”
 When we focus on efficiency, we care only about
outcomes, not about process
 But in real-world legal system, process is important in its
own right
48
Appeals
 In U.S., three levels of federal courts



District courts, circuit courts of appeals, Supreme Court
(Many state court systems also have three levels, but this varies by
state)
Parties in district court cases have right of appeal


Circuit court is required to consider their appeal
Parties in circuit court cases do not

Supreme Court has discretionary review – chooses which cases to
hear
 In common law countries, appeals courts tend to only
consider certain issues


Appeals generally limited to matters of law
Matters of fact generally not considered
49
Appeals
 Recall goal of legal system

Minimize administrative costs + error costs
 Clearly, appeals process increases administrative costs

So only efficient if it reduces error costs
 Reasons why appeals process may reduce error costs


Appeals courts are more likely to reverse “wrong” decisions than
“right” decisions…
…which leads to losing parties appealing more often when decision
was “wrong”
50