1. Econ 522 Economics of Law
Dan Quint
Spring 2013
Lecture 15

2. Reminders
HW3 (contract law) due Thursday

3. Last week Last week – lots of reasons a contract might not be enforced
Derogation of public policy
Incompetence, but generally not drunkenness
Duress and Necessity
Impossibility
Bad information (fraud, failure to disclose, frustration of purpose, mutual mistake), but generally not unilateral mistake
Vagueness, adhesion, unconscionability
For each: is there an economic justification?
Today: more on remedies for breach of contract
But first…

4. Results of Wednesday’s Experiment (trust)

5. The game we played Player A starts with $10
Chooses how much of it to give to player B
That money is tripled
Player B has $10, plus 3x whatever A gave him/her
Chooses how much (if any) to give back to player A
Tried it four ways:
Anonymous
On paper, but with names
Face to face
In public (in front of the class)

6. So, is trust a problem? With anonymity, yes
Average A sent $4.20, got back $4.66
Trust was rewarded, but just barely
Of those A’s who sent money, 27% got back less than they sent
Worse, less than half of the potential gains were realized
29% of A’s sent nothing, only 25% sent 10
With names – much less of a problem
Average A sent $6.95, got back $11.03, so 70% of gains realized
Only 8% of A’s sent nothing; of the rest…
13% got back less than they sent (10% got nothing)
49% got back at least twice what they sent
Face-to-face and public: every A sent $10, every B $15 or $20

7. So, is trust a problem? Anonymous With Names A sent # Obs Avg back
Avg gain
% less
% zero 20
0.00
+0.00
1-3 16
3.41
+1.16
19%
19%
Anonymous
4-6 15
5.40
+0.27
27%
27%
7-9 1
7.00
+0.00 0% 0% 10 17
10.53
+0.53
35%
24%
> 0
Avg: 5.92
6.56
+0.64
27%
22%
A sent
# Obs
Avg back
Avg gain
% less
% zero 6
0.00
+0.00
1-3 8
2.85
+0.38
38%
25%
With Names
4-6 18
9.19
+4.19 0% 0%
7-9 6
13.85
+6.51 0% 0% 10 36
15.14
+5.14
17%
14%
> 0
Avg: 7.56
12.01
+4.45
13%
10%

8. So, is trust a problem? Potentially, yes
In anonymous case, A didn’t trust B that much, and B wasn’t that trustworthy
Knowing each others’ names helped a significant amount…
…and interacting face-to-face solved the problem completely

9. Remedies for breach of contract
That’s all for ways to invalidate a contract. Next, we’ll consider the remedies that are available when a contract is breached.

10. Three broad types of remedy for breach of contract
Party-designed remedies
Remedies specified in the contract
Court-imposed damages
Court may decide promisee entitled to some level of damages
Specific performance
Forces breaching party to live up to contract
There are three general types of remedies for breach of contract:
party-designed remedies
court-imposed damages
specific performance
Party-designed remedies are remedies specified in the contract for a particular scenario
for example, a construction contract might stipulate a particular daily fee if completion of the building is delayed
Court-imposed damages are what we’ve already been talking about
the court may rule that the promisee is entitled to some amount of money as compensation for the breach
The third alternative is for the court to require specific performance
This forces the breaching party to live up to the contract
(Not always possible, of course
But if I agreed to sell you my painting, and then tried to breach the contract to sell it to someone else, the court could simply rule that I have to sell it to you at the agreed price.)
When a court awards monetary damages, there are still several different ways these damages could be calculated

11. Expectation damages
Compensate promisee for the amount he expected to benefit from performance
You agreed to buy an airplane for $350,000
You expected $500,000 of benefit from it
Expectation damages: if I breach, I owe you that benefit
($500,000 if you already paid, $150,000 if you didn’t)
“Positive damages”
Make promisee indifferent between performance and breach
We’ve already seen one: expectation damages.
Expectation damages are meant to compensate the promisee for the amount he expected to benefit from performance of the promise
Recall the airplane example we did a while back
you contracted to buy an airplane from me for $350,000
you expected to derive $500,000 of benefit from it
under expectation damages, I would owe you that benefit
($500,000 if you had already paid me, or $500,000 - $350,000 = $150,000 if you had not.)
The civil law refers to these as positive damages, as they compensate you for the positive benefit you anticipated from the contract
When they are calculated correctly, expectation damages make the promisee indifferent about whether the promisor performs or breaches
Thus, under expectation damages, the promisor internalizes all the costs of breach, and therefore makes the efficient decision about breach.

12. Reliance damages
Reimburse promisee for cost of any reliance investments made, but not for additional surplus he expected to gain
Restore promisee to level of well-being before he signed the contract
You contracted to buy the plane and built a hangar
If I breach, I owe you what you spent on the hangar, nothing else
“Negative damages” – undo the negative (harm) that occurred
A second type of damages are reliance damages
These compensate the promisee for any investments he made in reliance on the promise, but not for the additional surplus he expected to gain
Reliance damages, therefore, restore the promisee to the level of well-being he would have had if he had not received the promise in the first place
In the airplane example, suppose I chose not to deliver the plane but you had built yourself a hangar
Reliance damages would require me to reimburse you the cost of the hangar, but not the surplus you expected to earn from owning the plane
In the rich uncle example – the rich uncle promises his nephew a trip around the world, then changes his mind – reliance damages would pay for whatever supplies the nephew had purchased in preparation for the trip, minus whatever price he could resell them for
Again, this puts the promisee back in the position he was in before the promise was made
The civil law tradition refers to these as negative damages, as they undo the negative (the harm) that actually occurred in response to the promise
If no investments were made in reliance on the promise, reliance damages would be 0 – sort of a “no harm, no foul” rule.

13. Opportunity cost damages
Give promisee benefit he would have gotten from his next-best option
Make promisee indifferent between breach of the contract that was signed, and performance of best alternative contract
You value plane at $500,000
You contract to buy plane from me for $350,000
Someone else was selling similar plane for $400,000
By the time I breach, that plane is no longer available
I owe you $100,000 – the benefit you would have gotten from buying the other seller’s plane
A third type of damages are opportunity cost damages
If you agree to buy a plane from me, you may pass up another chance to buy a plane from someone else
But if I breach our contract, that other option may no longer be available
Opportunity cost damages are set to restore the promisee to the level of well-being he would have had if he had not contracted with this promisor, and instead had gone with his “next-best option”
Opportunity cost damages can be seen as an extension of reliance damages, where now turning down another opportunity is seen as a form of reliance, that is, as an investment you make in reliance on the promise that was made
Thus, opportunity cost damages leave the promisee indifferent between breach of the contract that was signed and fulfillment of the best alternative contract
In the airplane example, you contracted to buy a plane worth $500,000 for $350,000
Suppose someone else had an equally attractive airplane for sale at $400,000
If I breach our contract, opportunity cost damages would be $100,000, since this is the surplus you would have realized by foregoing the contract with me and instead buying the other plane

14. Example: expectation, reliance, and opportunity cost damages
You agree to sell me ticket to Wisconsin-Ohio State football game for $50
Expectation damages: you owe me value of game minus $50
If I pay scalper $150, then expectation damages = $100
Reliance damages: maybe 0, or cost of whatever pre-game investments I made
An example
Wisconsin plays Purdue in football in two weeks, I decide I want to go
One of you has a roommate with an extra ticket, and so you agree to sell it to me for $50
At the last minute, your roommate decides to go to the game, and you break your promise to me
Expectation damages are supposed to make me as well-off as if you had indeed sold me the ticket for $50
It may be hard to measure exactly how well-off the football game would have made me
But once you tell me my ticket is gone, I could show up at the stadium before the game and buy a ticket from a scalper
Say this costs $150
This gives us an easy way to compute expectation damages
If you had lived up to your promise, I’d have paid $50 to get into the game; since you broke your promise, I paid $150
So I’d be $100 better off if you’d kept your promise; so expectations damages are $100
Next, consider reliance damages for the same contract
Going to a football game doesn’t involve a lot of substantial investments
It’s possible reliance damages would be 0
If I had already gone out and bought red-and-white face paint and a big foam finger, reliance damages might reimburse me for these purchases
But reliance damages would not compensate me for the benefit I expected to get.

15. Example: expectation, reliance, and opportunity cost damages
You agree to sell me ticket to Wisconsin-Ohio State football game for $50
Expectation damages: you owe me value of game minus $50
If I pay scalper $150, then expectation damages = $100
Reliance damages: maybe 0, or cost of whatever pre-game investments I made
When you agreed to sell me ticket, other tickets available for $70
Opportunity cost damages: $80
(I paid a scalper $150 to get in; I would have been $80 better off if I’d ignored your offer and paid someone else $70)
Finally, suppose when we made the deal, there were lots of people offering tickets on Craigslist for $70
By the time you breach, these tickets were gone, so all I could do was pay a scalper $150 for a ticket
The actual payoff I got was G – 150, where G is the value of attending the game
If I had signed the best alternative contract, my payoff would have been G – 70
So while the contract we signed would have made me $100 better off, the best alternative would have made me $80 better off
Opportunity cost damages, then, would be set at $80, to compensate me for having passed on that opportunity.
Also note that these are all remedies for seller breach
We could calculate what I would owe you if I changed my mind and decided not to go to the game – that is, the remedy for buyer breach – in the analogous way.
In this example, a ticket to a football game is a good with many substitutes
That is, there are lots of tickets to the game, and they’re all worth about the same amount
So it made sense to calculate damages based on the market price of a replacement ticket
When a contract is for a unique good, this doesn’t always work; but conceptually, the analysis is almost the same.

16. ³ ³ = = = ³ ³ ³ ³ Ranking damages $100 $80 $15 Contract I Sign
Best Alternative
Do Nothing = = =
Breach + Expectation Damages
Breach + Opportunity Cost Damages
Breach + Reliance Damages
In this example,
Expectation Damages >= Opportunity Cost Damages >= Reliance Damages
It turns out, as long as all three are computed correctly, this should always be true
The reason is simple.
If I am rational and choose to sign a particular contract, it must be because that contract is at least as good for me as my best alternative
Of course, doing nothing is always an option, so it stands to reason that both the contract I sign, and the next best alternative, are at least as good as doing nothing
So from the point of view of my utility,
Contract I sign >= Best Alternative >= Doing Nothing
Following breach,
expectation damages restore me to the value of performance of the contract I signed
opportunity cost damages restore me to the value of performance of the next-best alternative
reliance damages restore me to the value of having done nothing So
Breach + Expectation Damages >= Breach + Opp Cost Damages >= Breach + Reliance D
If we subtract off the value of the breached contract in each case, we see that
(Of course, damages are not always calculated perfectly, so there may be instances in which this is violated. We’ll see examples of this next time.)
Expectation Damages
Opportunity Cost Damages
Reliance Damages
$100
$80
$15

17. Hawkins v McGee (“hairy hand case”)
Hawkins had a scar on his hand
McGee promised surgery to “make the hand a hundred percent perfect”
Surgery was a disaster, left scar bigger and covered with hair
In the examples so far – the airplane, and the football ticket – we could use market prices to calculate damages
In some situations, the value of a contract is subjective, making things a bit harder
Still, the principle is the same, it’s just a question of how to actually do the calculation.
A dramatic example of this is a 1929 case from New Hampshire, Hawkins v McGee, “the hairy hand case.”
George Hawkins had a scar on his hand from touching an electrical wire when he was young
A local doctor, McGee, approached him about having the scar removed, and promised to “make the hand a hundred percent perfect hand.”
Skin from Hawkins’ chest was grafted onto his hand
But the surgery was a disaster: the scar ended up bigger than before, and covered with hair
Hawkins successfully sued McGee; the issue on appeal was how high to set the damages.
To get a view of what’s going on, consider Hawkins’ preferences over two types of goods: his hand, and money. (Indifference curves.)

18. Hawkins v McGee (“hairy hand case”)$
+ Expectation Damages
+ Opp Cost Damages
Expectation Damages
Reliance Damages
Opp Cost Damages
+ Reliance Damages
To get a view of what’s going on, consider Hawkins’ preferences over two types of goods: his hand, and money. He could have a scar on his hand; a perfect hand; he could have gone to a different doctor who promised him improvement but not perfection; and he could have the hairy hand he had after surgery. Indifference curves.
Hawkins started out at the red dot – with a scarred hand, and some initial amount of money.
The doctor promised him the green dot – a perfect hand (but less money, since he would have to pay for the surgery)
Now he’s stuck with a hairy hand. And about to get more money (damages).
Reliance damages would get him back to the same level of well-being as before
The next-best doctor might have gotten him to the orange dot – a pretty good hand.
Opportunity cost damages would get him to the level of well-being he would have had by going to the next-best-doctor
Expectation damages would get him to the level of well-being he expected from the successful operation.
(Of course, there is still the question of how to calculate these indifference curves, since they are clearly subjective
There is also the question of whether it even makes sense to think that money can compensate for something like a disfiguring injury. But this is at least the principle.)
Once again, we see that the promised benefit (performance) is better than the next-best option, which is better than doing nothing
So expectation damages are bigger than opportunity cost damages, which are bigger than reliance damages
This ranking should hold whenever all three damage levels are calculated correctly
(The textbook gives examples of reasons that damages might be calculated incorrectly in real life – for example, if someone’s subjective valuation of a promise was very different from an objective, “market”-based measure – and in those cases, the order could be switched)
Initial Wealth
Hand
Hairy
Scarred
Next best doctor
100% Perfect

19. Other court-ordered remedies
Restitution
Return money that was already received
Disgorgement
Give up wrongfully-gained profits
There are three other types of damages that are sometimes ordered, which aren’t all that interesting, but are worth knowing.
Restitution is basically just forcing someone to pay back whatever money they’ve already received
I contract to buy a house and make a down payment
The seller breaches
If nothing else, he is required to return my down payment
This is sort of a minimal remedy, but it at least is very easy to implement.
Disgorgement is a similar concept, requiring someone to give up whatever profits they have wrongfully gained
Directors of a corporation have a duty to stockholders to act in a certain way
This is a very vague duty, not a list of specific things they are required to do
Suppose a director acts in a disloyal way, which is costly to the company but profitable to himself
Disgorgement damages would require the director to give up whatever profits he earned to the victim, in this case, the stockholders
Perfect disgorgement damages take away the director’s incentive to commit fraud

20. Other court-ordered remedies
Restitution
Return money that was already received
Disgorgement
Give up wrongfully-gained profits
Specific Performance
Promisor is forced to honor promise
Civil law: often ordered instead of money damages
Common law: money damages more common; S.P. sometimes used when seller breaches contract to sell a unique good
Like injunctive relief
Specific Performance is when, instead of ordering money damages, the court orders the promisor to live up to the promise
I contract to buy a particular house from you
It’s a beautiful house, in a historical neighborhood, and there are no others like it available
After we sign an agreement, you find a more eager buyer, and try to breach our agreement
The court might order you to follow through with the sale to me – this would be specific performance
In some civil law countries, specific performance rather than monetary damages is the traditional remedy for breach of contract
In common law, when a good has substitutes, money damages are traditionally the remedy, since the promisee can use that money to buy a substitute good if he so chooses
Specific performance is sometimes ruled for sale of goods without any substitutes
Specific performance can also be seen as an analog to injunctive relief in property law – your promise now belongs to me, so you’re not allowed to get out of it unless I choose to release you

21. Expectation damages vs. specific performance
Peevyhouse v Garland Coal and Mining Co (OK Supreme Court, 1962)
Garland contracted to strip-mine coal on Peevyhouse’s farm
Contract specified Garland would restore property to original condition; Garland did not
Restoration would cost $29,000
But “diminution in value” of farm was only $300
Original jury awarded $5,000 in damages, both parties appealed
OK Supreme Court reduced damages to $300
We said earlier that expectation damages will always lead to efficient breach
By equating liability with the social benefit of performance, we force promisor to internalize the externality of breach, so he breaches only when efficient
However, there is a case that illustrates how this can sometimes go wrong, when expectation damages aren’t appropriately measured
The case is Peevyhouse v Garland Coal and Mining Company (OK Sup Ct, 1962)
Peevyhouse owned a farm in Oklahoma
Garland contracted to strip-mine some coal on the property
The contract specified that Garland would take certain steps to restore the property to its previous condition after mining the coal
After mining the coal, Garland made no attempt to take these restorative steps
It was estimated at trial the restoration would cost $29,000
Peevyhouse sued for about that much ($25,000)
Both parties stipulated during trial that everything else in the contract had been performed
The defendant introduced evidence that although the damage would cost $29,000 to repair, it lowered the value of the plaintiffs’ farm by only $300
(The “diminution in value” of the farm due to the damage was $300.)
Original jury awarded Peevyhouse $5000 in damages; both sides appealed to OK Sup Ct
Peevyhouse saying this was less than the service that had been promised
Garland saying this was still more even than the total value of the farm after repairs
The OK Supreme Court reduced the damage award to $300.

22. Expectation damages vs. specific performance
At first, sounds like a perfect example of efficient breach
Performing last part of contract would cost $29,000
Benefit to Peevyhouses would be $300
Efficient to breach and pay expectation damages, which is what happened
But…
Most coal mining contracts: standard per-acre diminution payment
Peevyhouses refused to sign contract unless it specifically promised the restorative work
Dissent: Peevyhouses entitled to specific performance
(Peevyhouses seemed to value condition of property much more highly than change in market value)
At first glance, this seems like a nice example of efficient breach
Performing the last part of the contract would cost $29,000
The benefit to the Peevyhouses would be $300
So it is efficient to breach and pay expectation damages, which is what was awarded
However, the dissenting opinion noted that the coal company was well aware of what they were getting into when they signed the contract
Most coal mining contracts at that time contained a standard per-acre diminution payment, that the coal company would pay instead of repairing the property
The Peevyhouses specifically rejected that clause of the contract during initial negotiations, and would not sign the contract unless it specifically promised the restorative work
The dissent argued that the Peevyhouses were therefore entitled to “specific performance” of the contract, that is, to have the restorative work completed as promised.
Even though objectively, the damage to the property diminished its value by only $300, the Peevyhouses’ subjective value appears to have suffered much more
Expectation damages are meant to make the promisee as well off as they would have been under performance – here, this seems not to be the case.
At least one scholar has claimed that the judges who decided Peevyhouse were either incompetent or corrupt
one was later impeached, and others resigned
although others have disagreed
Still, it appears that the ruling here attempted to turn an efficient default rule – expectation damages – into a mandatory rule, that is, a rule that would be enforced even when it was not what the parties intended at the time of the contract.

23. Think about Peevyhouse in terms of penalty defaults
Contract promised restoration work, didn’t specify remedy if it wasn’t performed
Which default rule works better:
Default rule allowing Garland to breach and pay diminution fee?
Default rule forcing Garland to perform restoration work?
Ayres and Gertner: default rule should penalize the better-informed party
Garland routinely signed contracts like these
Peevyhouses were doing this for the first time
Default rule allows Garland to pay diminution fee: they have no reason to bring it up, Peevyhouses don’t know
Default rule forces Garland to do cleanup: if that’s inefficient, they could bring it up during negotiations
In this case, specific performance would serve as a penalty default

24. Party-designed remedies
Remedy for breach could be written directly into contract
But common law courts don’t always enforce remedy terms
Liquidated damages – party-specified damages that reasonably approximate actual harm done by breach
Penalty damages – damages greater than actual harm done
Civil law courts are generally willing to enforce penalty damages
But common law courts often do not
We mentioned that in some cases, the remedy for breach can be built directly into the contract
Some contracts stipulate an amount of money that a party would have to pay if he were to breach the contract
Others might specify a “performance bond” – an amount of money deposited with a third party, that would be paid to the promisee upon breach
Or a contract might specify a different process for resolving any disputes, such as agreeing to submit to binding arbitration or some other process
(This is what happens with disputes among diamond merchants, which are never resolved in court.)
For some reason, courts are often hesitant to enforce remedy terms in contracts
Rather than enforce the contract as written, courts sometimes set aside remedy clauses and impose their own remedies
One type of remedy that courts will often refuse to enforce is penalty damages
Penalty damages are damages that are greater than the actual harm that occurred
(“Liquidated damages” refer to party-specified damages that are a reasonable estimate of the actual harm done by the breach; penalty damages are any damages that are greater than that)
Civil law courts generally enforce penalty damages
But common law courts often set aside penalty damages, and only enforce liquidated damages

25. Penalty Damages Peevyhouse v Garland Coal
Coal worth $70,000 Garland to pay $25,000 Restoration would cost $30,000 Liquidated damages are $300 Peevyhouses value restoration at $40,000
Penalty Damages
Peevyhouse v Garland Coal
Peevyhouses only wanted farm strip-mined if it would be restored to original condition after
Suppose coal extracted worth $70,000
Garland paid $25,000 for rights to mine it
Restoration work would cost $30,000
Diminution of value was $300
So liquidated damages would be $300
Suppose Peevyhouses got $40,000 of disutility from land being left in poor condition
It’s unfortunate that penalty clauses are often not enforced, because they could be useful in some instances
In particular, they could be helpful when one party to a contract places a high subjective value on performance
Go back to coal mining example we saw earlier, Peevyhouse v Garland Coal
The Peevyhouses only wanted their farm strip-mined if the land could be restored to its original condition
They seemed to value the condition of their farm much more highly than “market value”
Had they known the coal company would refuse to perform the restoration work, and instead only pay them damages, they might have refused to agree to the contract in the first place
We can illustrate this with a simple dynamic game
Suppose the coal extracted was worth $70,000 (net of costs), and Garland Coal agreed to pay $25,000 to mine it
Recall that restoration work would have cost around $30,000, but damages for refusing to do it were assessed at $300
Suppose that the Peevyhouses got $40,000 of disutility from the land not being restored

26. Liquidated damages If damages limited to liquidated damages…
Coal worth $70,000 Garland to pay $25,000 Restoration would cost $30,000 Liquidated damages are $300 Peevyhouses value restoration at $40,000
Liquidated damages
Peevyhouses
Sign
Don’t
Garland Coal
(0, 0)
Restore property
Don’t, pay damages
(25,000, 15,000)
(-14,700, 44,700)
If damages limited to liquidated damages…
Peevyhouses shouldn’t believe restorative work will get done
So Peevyhouses better off refusing to sign
Even though mining and restoring Pareto-dominates

27. Penalty damages If penalty clauses in contracts were enforceable…
Coal worth $70,000 Garland to pay $25,000 Restoration would cost $30,000 Liquidated damages are $300 Peevyhouses value restoration at $40,000
Penalty damages
Peevyhouses
Sign
Don’t
Garland Coal
(0, 0)
Restore property
Don’t, pay penalty
One way to address this in a contract would have been to write into the contract a $40,000 penalty in the event that the restorative work was not completed
If this was enforceable, it would ensure that the mining company followed through with the restorative work
That is, it would change the game to this:
Now, the Peevyhouses are assured of getting their land restored
So they’re happy to agree to the contract
But, if common law courts don’t uphold penalty clauses, this won’t work
Another example: suppose you’re hiring a contractor to build a house, and you place a very high value on its being completed by a particular date
You are happy to pay $200,000 to get the house built, but insist on a $50,000 penalty if it’s not ready on that date
Now suppose there are lots of contractors who could potentially build the house
If you offer this deal to lots of them, the ones who accept are likely the ones who are most confident in their ability to finish the house on time
Since you value this highly, this may be efficient
And this may be the easiest (or the only) way for you to elicit this information
If there is no penalty clause, every contractor will try to convince you he’s 100% certain he’ll finish on time
But if someone accepts the $50,000 penalty clause, they’re probably pretty sure
(25,000, 15,000)
(25,000, 5,000)
If penalty clauses in contracts were enforceable…
Write contract with $40,000 penalty for leaving land unrestored
Now restoration work would get done, so Peevyhouses willing to sign
But if courts won’t enforce penalty damages, this won’t work

28. Penalty clauses
Whatever you can accomplish with penalty clause, you could also accomplish with performance bonus
I agree to pay $200,000 to get house built, but I want you to pay a $50,000 penalty if it’s late
Alternatively: I agree to pay $150,000 for house, plus a $50,000 performance bonus if it’s completed on time
Either way, you get $150,000 if house is late, $200,000 if on time
Courts generally enforce bonus clauses, so no problem!
So in some instances, penalty damages seem beneficial
Especially if one party would not agree to a mutually beneficial contract without them
So it seems like the fact that they’re not enforced is a problem
However, it turns out, most things you could accomplish with penalties, you could restate in a different way with a performance bonus
Go back to the house-building example
I’m happy to pay $200,000 to get the house built by a certain date
But I insist on a $50,000 penalty if it’s not ready on time
We could alternatively write the contract this way:
I pay $150,000 for the building of the house
And I pay an additional $50,000 performance bonus if the house is completed by a certain date
Courts generally have no problem enforcing contracts with bonuses in them, so this would likely be enforced as written
But it has the same effect as the contract with the penalty
(Under either contract, the builder gets $200,000 for finishing the house on time, and $150,000 for finishing it late.)

29. Penalty clauses
Whatever you can accomplish with penalty clause, you could also accomplish with performance bonus
I agree to pay $200,000 to get house built, but I want you to pay a $50,000 penalty if it’s late
Alternatively: I agree to pay $150,000 for house, plus a $50,000 performance bonus if it’s completed on time
Either way, you get $150,000 if house is late, $200,000 if on time
Courts generally enforce bonus clauses, so no problem!
Similarly, Peevyhouse example
Peevyhouses get $25,000 for mining rights, $40,000 penalty if land is not restored
Equivalently, get $65,000 for mining rights, pay $40,000 bonus if restoration is completed
But, if intent of contract is too transparent, still might not be enforced
Similarly, go back to the Peevyhouse example
Instead of a $40,000 penalty that might not be upheld, the Peevyhouse contract also could have been rewritten as a bonus
We imagined that Garland paid $25,000 to mine the coal
Suppose the contract were written in this way:
Garland pays the Peevyhouses $65,000 to mine the coal
The Peevyhouses pay Garland a $40,000 bonus if the resoration work is completed
This would create exactly the same incentives as a $40,000 penalty clause
(In this case, though, the intent of the contract might have been so transparent that it still might not have been enforced)

30. Effects of different remedies on…. decision to perform or breach
Effects of different remedies on… decision to perform or breach decision to sign or not sign investment in performing investment in reliance
Now, once all the decisions have been made – once we’ve decided to sign a contract, you’ve decided how much to rely, I’ve decided to breach, and the case is in front of a judge or jury – the remedy for breach of contract is purely a question of redistribution
So if we want to understand the impact of remedies on efficiency, we can think only about the incentives they create for our decisions, specifically…

31. Costs High – Renegotiate
Plane worth $500,000 to you Price $350,000 Cost: either $250,000 or $1,000,000
Remedies and breach
Expectation Damages
Specific Performance
Costs Low – Perform
Costs High – Perform
Costs High – Breach
Costs Low – Perform
Costs High – Perform
Costs High – Renegotiate
I get
100,000
-650,000
-150,000
I get
100,000
-650,000
-400,000
–650, ½ (500,000)
You get
150,000
150,000
150,000
You get
150,000
150,000
400,000
150, ½ (500,000)
For example, let’s compare a contract enforced by specific performance – the promisor is not allowed to breach unless the promisee agrees – versus a promise enforced by expectation damages
Let’s stick with the example of the airplane I agreed to build for you
The plane will give you a value of $500,000; we agree on a price of $350,000
I expect the plane to cost me $250,000 to build, but there is some chance it will instead cost me $1,000,000
Expectation damages
Specific performance
by renegotiating the contract, we create a total surplus of $500,000 (beyond our outside options if you forced me to build the plane)
Suppose that our negotiations lead us to divide this additional surplus evenly
This would lead you to get $250,000 beyond your outside option, which was $150,000, so your payoff would be $400,000
I would get $250,000 beyond my outside option, which was -650,000, which would be -400,000
(So this means, in return for getting out of the contract, I pay you $400,000)
As long as transaction costs are low, either remedy will lead to the same outcome
When it is efficient to breach, expectation damages lead me to breach without permission
Under specific performance, when it is efficient to breach, renegotiation will lead us to some sort of agreement to let me off the hook.
But when the transaction costs of renegotiating our contract are too high…
Expectation damages still allow me to breach
Specific performance would lead to inefficient performance
I would have to build you the plane, even though it costs me more than your benefit
Total
250,000
-500,000
Total
250,000
-500,000
Transaction costs low  either leads to efficient breach, but seller prefers “weaker” remedy
Transaction costs high  S.P. leads to ineff. performance

32. Remedies and breach Opportunity cost damages, or reliance damages
Inefficient breach when transaction costs are high
Renegotiate contract to get efficient performance when transaction costs are low
Like nuisance law: any remedy leads to efficient breach with low TC
But only expectation damages do when TC are high
Unfortunate contingency and fortunate contingency
We can do the same type of analysis with other types of damages – opportunity cost damages, or reliance damages
With high transaction costs, either of these would lead to inefficient breach – sometimes, I would choose to breach, even when that’s inefficient
With low transaction costs, when performance is efficient but I would otherwise choose to breach, we can simply renegotiate the contract and get performance
So just like in nuisance law, any remedy leads to efficient performance when transaction costs are low
But only expectation damages do it without renegotiation; and only expectation damages get efficient breach even when transaction costs are high
An increase in the cost of performance is referred to as an unfortunate contingency
On the other hand, I may want to break my promise because I discover another buyer who values my product more than you do
This is referred to as a fortunate contingency
Earlier, we saw the example where you contracted to buy my painting, but my rich cousin appeared and valued it much more highly than you did
We can do the same exercise as before with the different remedies, and see the same thing
With low transaction costs, any remedy will lead to efficient breach, but with different allocations of surplus
An example of this is done in the textbook (p. 267)
Once again, specific performance is the most advantageous to the buyer, and higher damages are better for the buyer than lower damages.

33. Efficient signing Specific Performance
If costs stay low, I get $350,000 – $250,000 = $100,000 profit
If costs rise, I take $400,000 loss
Am I willing to sign this contract?
Even expectation damages face this problem
Expectation damages: costs stay low, same $100,000 profit
Costs rise, $150,000 loss
If probability of high costs is ½, I won’t sign contract
Expectation damages lead to efficient breach, but may not lead to efficient signing
One further point that we mentioned earlier but didn’t talk much about: efficient signing.
With specific performance, in the example above, we saw that when costs stay low and I perform, I get a profit of $100,000 (the price you pay, $350, minus the costs I incur, $250)
When costs rise and I need to renegotiate to get out of the contract, if gains from trade were split evenly, I would take a $400,000 loss
Which begs the question: would I have been willing to sign this contract in the first place?
It turns out, expectation damages face this problem too
When my costs remain low and I perform, I get the same profit of $100,000
And when my costs jump up and I breach, I owe $150,000 (expectation damages)
If the probability my costs rise is small enough, that’s no big deal
I take the risk of owing you expectation damages, because my profits in the cases where I don’t breach are large enough that it’s worth the risk.
But if the probability my costs rise is high, then my expected profit from this contract is negative
For example, suppose the probability of a dramatic rise in costs is ½
Then my expected payoffs from agreeing to build you the airplane are
½ (100,000) + ½ (-150,000) = -25,000
So now there’s no way I’d agree to the contract in the first place
This is the point we mentioned before:
Expectation damages lead to efficient breach, but they may lead to inefficient signing
(If I’m the only airplane manufacturer available, it’s still efficient for us to sign this contract – it generates positive expected surplus – but under expectation damages, I would not sign this contract.)
This suggests that, even if expectation damages make a sensible default rule, it’s efficient for parties to be able to specify a different damages rule in the contract
That is, even if expectation damages are often efficient, they are not always efficient, so there’s no reason for them to be mandatory

34. Reliance – did example a while ago
If reliance investments increase the damages you receive, we expect to get overreliance
To get efficient reliance, need to exclude gains from reliance in calculation of expectation damages
But then promisor’s liability < promisee’s benefit, leading to inefficient breach
With low transaction costs, fix this through renegotiation
But what about unobservable actions the promisor needs to take, to make breach less likely?
Investment in performance
So if damages include the gains from reliance, this leads to overreliance
But if the level of damages excludes the gains from reliance, this leads to efficient reliance
But we also know that if the level of damages exclude the gains from reliance, and the buyer relies anyway, that this will lead to inefficient breach
That is, since the seller’s liability from breach is lower than the buyer’s gain from performance, there will be some instances where the seller breaches even though efficiency requires performance
We said before that with no transaction costs, this problem can be solved
When breach would be inefficient, the parties can contract around it
Since it’s generally very clear whether a party has breached a contract or not, this shouldn’t be a particular problem.
However, there are situations in which a promisor can take actions to make performance less costly
Or, to put it another way, to lessen the probability that breach is necessary
A contractor could buy raw materials ahead of time, to avoid the risk of changing prices
A manufacturer could start a project earlier and frontload the labor to avoid the risk of a strike
If a project were to require a building permit or zoning easement, he could lobby the local government (or bribe someone) to decrease the chances of hitting a snag.
This sort of investment in performance, however, may be unobservable, or unverifiable
So even when this sort of investment is efficient, it may be very hard to build it into the contract
(It’s very hard to enforce a contract specifying “how hard I have to try” to convince the zoning board to approve your project.)
So investment in performance may only occur when it is in the promisor’s best interest
Skip

35. Investment in performance
Some investment I can make to reduce likelihood that breach becomes necessary
Suppose probability of breach is initially ½…
but for every $27,726 I invest, I cut the probability in half
Invest nothing  probability of breach is 1/2
Invest $27,726  probability is 1/4
Invest $55,452  probability is 1/8
Any investment z  probability is .5 * (.5) z / 27,726
Wrote it this way so p = .5 e – z / 40,000
We’ll stick with the airplane example
Suppose there is some investment I can take that reduces the chance that my costs go up
I can buy some of my supplies ahead of time
This will reduce the risk of having to breach
But I can’t completely insulate myself from risk
Suppose that for each $27,726 I invest, I can reduce the probability I will need to breach by ½
That is, if I invest $27,726, the probability of breach goes from ½ to ¼
If I invest another $27,726, it goes down to 1/8
And so on
For any investment z, then, we can write the probability that breach is still required as ½ * ½ ^(z / 27,726)
The reason I chose the number 27,726 is that we can rewrite this probability as
½ e^{- z/40,000)
which makes it much easier to work with.

36. Investment in performance (continuing with airplane example)
Suppose you’ve built a $90,000 hangar
Increases value of performance by $180,000…
…so value of performance is $150,000 + $180,000 = $330,000
Probability of breach = .5 e – z/40,000
Let D = damages I owe if I breach
Same questions as before:
What is efficient level of investment in performance?
How much will I choose to invest in performance?

37. Investment in performance (continuing with airplane example)
Suppose you’ve built a $90,000 hangar
Increases value of performance by $180,000…
…so value of performance is $150,000 + $180,000 = $330,000
Probability of breach = .5 e – z/40,000
Let D = damages I owe if I breach
Same questions as before:
What is efficient level of investment in performance?
Enough to reduce probability of breach to 40,000/430,000
How much will I choose to invest in performance?
Enough to reduce probability of breach to 40,000/(100,000 + D)
Efficient level?
Social surplus is 330, ,000 if costs stay low, 0 if costs rise
So expected social surplus is (1 – p(z)) (430,000) – z
430,000 – 430,000 * 0.5 * e–z/40,000 – z
Take derivative to maximize this:
– 430,000 * 0.5 * (–1/40,000) e– z/40,000 – 1 = 0
Could solve for z, but simpler to solve for p(z)
0.5 * e– z/40,000 = 40,000 / 430,000 = p(z)
Efficient level of investment in performance is enough to reduce probability of breach to 40,000/430,000
What will promisor actually do?
Private gain to promisor is 100,000 when costs stay low, -D when costs go up, minus whatever he invests in performance
Expected private gain is
(1 – p(z)) 100,000 + p(z) (-D) – z
100,000 – p(z) (100,000 + D) – z
100,000 – (100,000 + D) * 0.5 * e–z/40,000 – z
– (100,000 + D) * 0.5 * (–1/40,000) e– z/40,000 – 1 = 0
Again, solve for p(z)
0.5 * e– z/40,000 = 40,000 / (100,000 + D) = p(z)
For any level of liability D, promisor invests enough to reduce probability of breach to 40,000/(100,000+D)

38. What do these results mean?
What is the efficient level of investment in performance?
Enough so that p(z) = 40,000/430,000
What will promisor do under various rules for damages?
Enough so that p(z) = 40,000/(100,000 + D)
So if D = 330,000, efficient investment in performance
D = 330,000 is promisee’s benefit, including reliance
So expectation damages, with benefit of reliance, leads to efficient investment in performance
If D < 330,000, too little investment in performance
If D > 330,000, too much
Makes sense – think about externalities
The efficient level of investment leads to a 40,000/430,000 chance of breach
Damages which are set equal to D lead to a self-interested promisor to allow a probability of breach of 40,000/(D + 100,000)
So when damages are set to 330,000 – damages include the benefit from reliance – the investment in performance is efficient
When damages are lower than that, the seller will underinvest in performance, leaving the risk of breach inefficiently high
When damages are higher than this, the seller will overinvest in performance
And what’s the overall lesson?
Making the seller liable for reliance – that is, increasing expectation damages to include benefit due to reliance – leads the seller to invest efficiently in performance; just like it led to efficient breach
But it leads the buyer to overinvest in reliance
On the other hand, making the seller not liable for reliance – leaving expectation damages where they were without reliance – leads to underinvestment in performance
But it leads to the efficient level of reliance
So like we saw last week with the sailboat example from Friedman:
The level of damages leads to multiple different incentives
And it’s impossible to come up with a level of damages that makes everyone behave efficiently
We already saw one attempt to get around this problem:
limit damages to the efficient level of reliance, not the actual level of reliance
that way, the promisee has no incentive to overinvest in reliance, so he invests the efficient amount
but the promisor still internalizes the full cost of breach
so he invests the efficient amount in performance
Textbook gives another clever, if unrealistic, solution, which is where we’ll start on Thursday

39. Effects of different remedies on…. decision to perform or breach
Effects of different remedies on… decision to perform or breach decision to sign or not sign investment in performing investment in reliance

40. Paradox of compensation
Expectation damages include benefit from reliance investments
Expectation damages exclude benefit from reliance investments
Efficient breach
Efficient investment in performance
Over-reliance
Inefficient breach
Underinvestment in performance
Efficient reliance
Summing up most of these results, we found that
When expectation damages include the anticipated benefit from reliance, we get
efficient breach, and efficient investment in performance
seller internalizes the full cost of breach, and makes efficient decision
inefficiently high reliance
buyer reliance imposes a negative externality on seller – since he has to pay more if he breaches – so buyer does not bear full “cost” of reliance
On the other hand, when expectation damages exclude the anticipated benefit from reliance,
inefficient breach/investment in performance
breach imposes a negative externality on buyer, so seller does not bear full cost
efficient reliance
buyer bears full cost, full benefit of reliance, makes efficient decision
This is what Cooter and Ulen call the paradox of compensation
neither rule gets both parties to act efficiently
Is there a way to get efficient behavior by both parties?
Skip

41. We already saw one possible solution
Have expectation damages include benefit from reliance…
…but only up to the efficient level of reliance, not beyond
That is, have damages reward efficient reliance investments, but not overreliance
Promisee has no incentive to over-rely  efficient reliance
Promisor still bears full cost of breach  efficient performance
Problem: this requires court to calculate efficient level of reliance after the fact
We already saw one proposed solution to this problem
Modify expectation damages to include the benefit of reliance, but only up to the efficient level of reliance, not beyond
This way, buyer has no incentive to over-rely, since he won’t get that benefit in the event of breach
And seller still bears full cost of breach, so he makes efficient decisions as well

42. Another clever (but unrealistic) solution
The problem:
Damages promisor pays should include gain from reliance if we want to get efficient performance
Damages promisee receives should exclude gain from reliance if we want to get efficient reliance
Solution: make damages promisor pays different from damages promisee receives!
How do we do this? Need a third party

43. “Anti-insurance” You (promisee) and I (promisor) offer Bob this deal:
If you rely and I breach,
I pay Bob value of promise with reliance (airplane plus hangar)
Bob pays you value of promise without reliance (airplane alone)
Bob keeps the difference
You receive damages without benefit from reliance; I pay damages with benefit from reliance
How do we do this?
Let’s say I’m the promisor, and you’re the promisee, and I’ve promised to build you an airplane
You and I have this friend, Bob
Bob likes money
So we go to Bob and say, hey Bob, here’s a deal for you
I’m planning to build a plane
He’s planning to buy the plane
He’s probably going to want to build a hangar
I might end up not building the plane
Here’s what we need you to do
In the event that he builds a hangar and I don’t build the plane,
I’m going to give you the value of the plane with the hangar;
And you’re going to give him the value of the plane without the hangar
And you keep the rest for yourself. OK?
And Bob says, “cool!”
This is called “anti-insurance”
Rather than buying insurance from a third party, you and I are entering into this additional contract where if things go bad, I owe Bob some additional money, beyond what I pay you
By doing this, we set both our incentives correctly
You only get the benefit of reliance if I perform, so you invest the efficient amount in reliance
I face the full cost of breach, so I invest the efficient amount in performance.

44. “Anti-insurance” You (promisee) and I (promisor) offer Bob this deal:
If you rely and I breach,
I pay Bob value of promise with reliance (airplane plus hangar)
Bob pays you value of promise without reliance (airplane alone)
Bob keeps the difference
You receive damages without benefit from reliance; I pay damages with benefit from reliance
Offer the deal to two people, make them pay up front for it
Now obviously, Bob is happy to do this for free; but it makes you and me worse off
So now we go to our other friend Carol, and say, Hey, Carol
Here’s a deal we’re offering
Give us $5 now, and if he builds a hangar and I don’t deliver a plane, you’ll get the difference between the value of the plane with the hangar and without the hangar
And Carol realizes this is worth more than $5, so she says, “sure.”
But now we go back to Bob, and we offer him the deal at $10 instead.
And if we make Bob and Carol compete for this deal, we should be able to get them to pay a fair amount for it up front
If they’re risk averse, of course, they’ll need to be compensated for taking on some risk
But if we have a risk-neutral friend who’s smart enough to understand the probabilities and figure out what each of us will do given our incentives, we can get them to give us the full value of the anti-insurance deal ahead of time, and divide it up among ourselves
So this way, we can give ourselves incentives for efficient reliance and efficient investment in performance at the same time.
So that’s another way to solve the paradox of compensation – even if it’s pretty unrealistic

45. Reminder: what do courts actually do?
Foreseeable reliance
Include benefits reliance that promisor could have reasonably anticipated
What courts actually do, rather than basing damages on actual or on efficient reliance, is to base damages on foreseeable reliance
That is, they base damages on what the promisor could have reasonably expected the promisee to do, not what he actually did
This was the decision in Hadley
Since the shipper could not reasonably expect the miller to rely so heavily, he was not liable for the lost profits
Most millers at the time had more than one crankshaft
So a broken shaft would not typically lead to shutdown
Under the doctrine of foreseeable reliance, if Hadley had told Baxendale that his mill was closed until repairs were made, then Baxendale would be liable for lost profits due to delay
By informing Baxendale of the reliance, Hadley would have made it foreseeable, and therefore compensable

46. Repeated interactions (we won’t get to this)

47. Repeated games
Nearly everything we’ve done so far has assumed a one-shot interaction
that is, we’ve been assuming that the parties to a contract are only interested in maximizing their gain from that one particular contract,
and are not concerned with any future interactions with the same partner.
Of course, in many cases, this is not true
Example. There’s a coffee shop near my house, and I go there several times a week.
One day, I forget my wallet, and ask if I can still buy a cup of coffee and a muffin, and pay them back the next day
They say yes, and I show up the next day with the money. Why?
(Seems pretty obvious to any reasonable people, less so to an economist.)
An economist would say:
The reason I pay them back is that I want to keep transacting with them in the future
And the value I expect to get from those future transactions is worth more to me than the $3 I could save by breaking my promise now
And the reason they trusted me is that they expected this would be the case.
From a theoretical point of view, repeated games can be very hard to analyze, because a lot of different things can happen
But one of the things that can happen is that we can cooperate in a repeated game, even if we could not cooperate in the same one-shot game.

48. Repeated games Suppose we’ll play the game over and over
Player 1 (you)
Trust me
Don’t
Player 2 (me)
(100, 0)
Share profits
Keep all the money
(150, 50)
(0, 200)
Let’s go back to the original agency game we did a couple weeks ago.
You choose whether to trust me with $100, which I can double by investing it; and then I decide whether to keep the $200 or return $150 to you and keep $50 for myself
But now, suppose there is the possibility of playing the game more than once.
In particular, suppose that each time we play the game, there is a 10% chance it’s the last time we play, and a 90% chance that we get to play again.
Suppose we’ll play the game over and over
After each game, 10% chance relationship ends, 90% chance we play at least once more…

49. Repeated games Suppose you’ve chosen to trust me
Keep all the money: I get $200 today, nothing ever again
Share profits: I get $50 today, $50 tomorrow, $50 day after…
Value of relationship =
Since this is more than $200, we can get cooperation
Think about my incentives to repay your money or keep it for myself
If I keep it for myself, I get a payoff of $200; but then you’ll never trust me again, so that’s all I’ll ever get
On the other hand, if I give you back your $150, you’ll probably trust me again the next time, and the time after that, and the time after that (provided I keep returning it)
So the value I expect to get out of the relationship is
X ^2 X ^3 X 50 + … = 50 / (1 - .9) = 500 > 200
So I’m much better off returning your money, since I’ll make more money in the long-run if you keep trusting me
And because of this, it makes sense for you to trust me

50. Repeated games Suppose you’ve chosen to trust me
Keep all the money: I get $200 today, nothing ever again
Share profits: I get $50 today, $50 tomorrow, $50 day after…
Value of relationship =
Since this is more than $200, we can get cooperation
The same thing can also happen with a repeated version of the prisoner’s dilemma.
Recall that in the one-shot prisoner’s dilemma, the only equilibrium was for both of us to rat on each other and go to jail for several years
However, if we expect to play the prisoner’s dilemma over and over, it turns out to be an equilibrium to both keep quiet
Actually, what turns out to be an equilibrium is for both of us to do the following:
Keep quiet the first time we play
As long as neither of us has ever ratted on the other, keep quiet
Once either of us has ever ratted, I rat every time we play forever
This is called a “grim trigger” strategy – we play a good (cooperative) strategy, but if either of us every rats, this triggers a “punishment phase” where we are unable to cooperate
The threat of moving to this punishment phase keeps us both quiet, even though either of us would gain in the short-term by ratting each other out.
So in the prisoner’s dilemma, or the agency game, if the game will be played over and over, it’s possible to get cooperation.
Similarly, in situations where contracts cannot be enforced, repeated interactions with the same parties can lead to voluntary cooperation.
Even when you won’t always be transacting with the same person, transacting within a small community, where people are aware of your reputation, leads to a similar incentive.

51. Repeated games and reputation
Diamond dealers in New York (Friedman)
“…people routinely exchange large sums of money for envelopes containing lots of little stones without first inspecting, weighing, and testing each one”
“Parties to a contract agree in advance to arbitration; if… one of them refuses to accept the arbitrator’s verdict, he is no longer a diamond merchant – because everyone in the industry now knows he cannot be trusted.”
The Friedman book mentions an article by Lisa Bernstein on diamond dealers in New York. Quoting from Friedman:
Buying and selling diamonds is a business in which people routinely exchange large sums of money for envelopes containing lots of little stones without first inspecting, weighing, and testing each one.
The New York diamond industry was at one time dominated by orthodox Jews, forbidden by their religious beliefs from suing each other – making it a trust-intensive industry conducted almost entirely by people who could not use the legal system to enforce their agreements. While the industry had become more diverse by the time Bernstein studied it, dealers continue to rely almost entirely on private mechanisms to enforce contracts – in part for religious reasons, in part to maintain privacy, in part, perhaps, because those mechanisms functioned better than the courts.
At the center of the system is the New York Diamond Dealers’ Club, which arranges private arbitration of disputes among diamond merchants. Parties to a contract agree in advance to arbitration; if, when a dispute arises, one of them refuses to accept the arbitrator’s verdict, he is no longer a diamond merchant – because everyone in the industry now knows he cannot be trusted. Similar arrangements exist elsewhere in the world and exchange information with each other. Presumably the amount diamond merchants are willing to risk on a single deal depends in part on how long the other party has been involved in the industry and thus how much he would lose if he had to leave it.

52. Repeated games and reputation
The first purpose of contract law is to enable cooperation, by converting games with noncooperative solutions into games with cooperative solutions
The sixth purpose of contract law is to foster enduring relationships, which solve the problem of cooperation with less reliance on courts to enforce contracts
Law assigns legal duties to certain long-term relationships
Bank has fiduciary duty to depositors
McDonalds franchisee has certain duties to franchisor
Recall that in the first couple lectures on contract law, we introduced a number of pronouncements from Cooter and Ulen, including:
The first purpose of contract law is to enable cooperation, by converting games with noncooperative solutions into games with cooperative solution
Considering the effect of repeated games, Cooter and Ulen add a sixth one to the list:
The sixth purpose of contract law is to foster enduring relationships, which solve the problem of cooperation with less reliance on the courts to enforce contracts.
We saw with the agency game that one way to allow for cooperation was to introduce enforceable contracts
Repeated interactions, or enduring relationships, give another way to get cooperation in what seems like a game with a noncooperative solution
The textbook gives a couple of examples of how courts sometimes try to foster enduring relationships:
By assigning legal duties to relationships that arise out of contracts
example: a bank has a fiduciary duty to its depositors which goes well beyond the terms they agree to on the account
a franchisee who runs a local McDonalds has certain duties to the franchisor
These duties are meant to encourage an enduring business relationship.
Similarly, courts sometimes treat long-term business relationships differently
encouraging the parties to “repair the relationship” rather than simply ruling on the merits of the dispute.

53. Repeated games and the endgame problem
Suppose we’ll play agency game 60 times
$50 x 60 = $3,000 > $200, so cooperation seems like no problem
But…
In game #60, reputation has no value to me
Last time we’re going to interact
So I have no reason not to keep all the money
So you have no reason to trust me
But if we weren’t going to cooperate in game #60, then in game #59…
So we’ve seen that repeated interactions and reputation can solve the agency problem and lead to cooperation
But there’s also a potential problem with this
In the examples so far, we’ve assumed that we don’t know how long we’ll keep interacting, but that it might go on indefinitely
However, if there is a particular date when we know our relationship will be over for sure, this can lead to a problem.
Suppose we are going to play the agency game once a month for five years – that is, we’re going to play 60 times
Clearly, getting $50 60 times is much better than getting $200 once, so it seems there should be no problem with getting me to return your money early on in the interaction.
However, think about the problem from the other end.
The last time we play, there’s no longer any future gain if I return your money, since we already know it’s our last time playing.
Given that, you expect me to steal the money you give me in game #60, so you don’t lend me the money.
But now think about game #59
We both know that we won’t cooperate in game #60 – that’s true regardless of what happens in game #59.
So what happens in game #59 doesn’t affect future play
So once again I have no reason to return your money, so I keep it; and you expect this, so you don’t trust me.
But now if we don’t expect to cooperate in game #59 or #60, there’s no reason for you to trust me in game #58.
And so on.
So in a finitely repeated game, cooperation can unravel from the back, and we can fail to cooperate form the very beginning!
This is referred to in Cooter and Ulen as the “endgame problem”
Once we know there is a finite end date to our interaction, cooperation can unravel

54. Repeated games and the endgame problem
Endgame problem: once there’s a definite end to our relationship, no reason to trust each other
Example: collapse of communism in late 1980s
Communism believed to be much less efficient than capitalism
But fall of communism led to decrease in growth
Under communism, lots of production relied on gray market
Transactions weren’t protected by law, so they relied on long-term relationships
Fall of communism upset these relationships
This is referred to in Cooter and Ulen as the “endgame problem”
Once we know there is a finite end date to our interaction, cooperation can unravel
They discuss an example where this happened: the collapse of communism across much of eastern Europe in 1989.
Communism was believed to be much less efficient than capitalism
But when central planning was replaced by markets, this actually led to a decrease in growth rather than the anticipated increase. Why?
Under communism, a lot of production relied on the black market, or the semi-legal gray market.
Since these transactions weren’t protected by law, they relied on long-term relationships to accomplish cooperation.
However, the fall of communism, and the uncertainty that came with it, upset these relationships, causing lots of important cooperation to break down.
So while repeated interaction gives us a way to cooperate without relying on courts to enforce contracts, having a definite end date to the interaction can spoil it.
Keep in mind, though, that to sustain cooperation, the game doesn’t have to actually go on forever; it just has to be possible at each stage that it might continue.
The probability the game will be played again is like a discount factor – you discount the future gains by the probability they will occur
As long as this probability is not too low, the possibility of future gains may still be enough to sustain cooperation.

55. One other bit I like from Friedman

56. Friedman on premarital sex
In most societies throughout history, people have wanted to have sex, often without getting married
But until recently, in most societies, men have wanted to marry women who have never slept with anyone else
(Could be evolved instinct to make sure kids are yours; could be a cultural/religious thing; but it’s historically true)
Creates a problem: single women are reluctant to have sex for fear it will lower their ability to find a suitable husband; and so single men have trouble finding women to sleep with.
One traditional solution is for unmarried couples to sleep together on the understanding that if the woman gets pregnant the man will marry her.
This practice was sufficiently common in a number of societies… that between a quarter and half of all brides went to the alter pregnant.
Of course, there is still the problem of opportunistic breach by the man
There are a variety of different ways to try to enforce this understanding
But historically, one way has been through contract law
Under traditional common law, a jilted bride could sue for breach of promise to marry.
The damages she could collect reflected the reduction in her future marital prospects.
They were in fact, although not in form, damages for loss of virginity.
However, starting in the 1930s, U.S. courts became increasingly unwilling to enforce this rule; by 1945, half the U.S. population lived in states where breach of promise to marry was no longer recognized as a legal issue
Which created a problem for women who wanted to have premarital sex, but did not want to end up single mothers in a society in which that status was both economically difficult and heavily stigmatized.

57. Friedman on premarital sex
Enter a clever new solution: the engagement ring
Contrary to what DeBeers may tell you, the tradition of a man giving his fiancee a valuable engagement ring is not an ancient custom
Friedman cites an article by Margaret Brinig, called “Rings and Promises”. She looks at diamond import data and finds that the practice appeared to became common only in the 1930s, peaked in the 1950s, and has since declined.
Her explanation was that the engagement ring served as a performance bond for the promise to marry.
Instead of suing, which was no longer an option, an abandoned bride could simply keep the ring, confiscating the posted bond.
The practice eventually declined not because of further legal changes – today, no states recognize the action for breach of promise to marry – but as a result of social changes.
As premarital sex became more common, contraception more reliable, and virginity of less importance on the marriage market, the risk of opportunistic breach, and thus the need for a bonding mechanism, declined.

58. That’s it for contract law
Purposes for contract law:
Encourage cooperation
Encourage efficient disclosure of information
Secure optimal commitment to performance
Secure efficient reliance
Provide efficient default rules and regulations
Foster enduring relationships
Next week, we begin tort law
The first purpose for contract law is to encourage cooperation, by changing games with noncooperative equilibria into games with cooperative equilibria
Encouraging cooperation means that contracts should generally be enforceable when both parties wanted them to be enforceable
The second purpose is to encourage efficient disclosure of information
This is sometimes accomplished through penalty defaults – rules that penalize a better-informed party that withheld information
The third purpose is to secure optimal commitment to performance – that is, efficient investment in performance, and breach only when it’s efficient
This is generally accomplished through expectation damages
The fourth purpose is to secure efficient reliance investments
This requires not including the anticipated benefits from overreliance in damages – but we saw multiple ways to try to accomplish this, and the paradox of compensation
The fifth is to provide efficient default rules and regulations
Efficient default rules generally require allocating a risk to the low-cost bearer
And we saw a number of examples of regulations – situations where contracts would not be enforced
The sixth purpose is to foster enduring relationships – as we saw today, repeated interactions and reputations can lead to cooperation without relying on the court system, but endgames pose a problem
End of material on second midterm