Hadley v. Baxendale
What was the contract? To carry a millshaft from Glouster to Greenwich. What was the breach? A delay; it got there six days late (because it went by canal boat part way instead of the train). Was this really a breach? The contract does not specify time for delivery, but let's assume that the delay was breach. Then we have an enforceable promise that was not kept. The question is: What are the damages?
Expectation damages in Hadely
Where are the Hadely brothers after the breach? They lost six days operation of the mill in addition to what they would have lost had the shaft been delivered on time.
Was there any way to mitigate those damages? Any way to get another shaft sooner? Apparently not.
Where would they have been had the contract been performed? They would have only lost a few days operation.
So what do they need to be put in as good a position as they would have been in had the contract been performed as promised? Six days lost profit.
Any difficulty in proving the amount of this profit? Apparently not.
Did the court give lost profit? No. Why not?
A new limit on expectation damages
The court quotes the expectation rule, but then says that "the proper rule in a case such as this" is different. What is this different rule? You only get the damages that may be reasonably supposed to be within the contemplation of the parties at the time of contracting; alternatively, the damages that would arise in the normal course of things. This is usually described as a foreseeability test.
One thing is clear: Courts are saying there is some sort of limit on the right to receive expectation damages. There are some cases in which courts deliberated award less than expectation damages.
Hadley's foreseeability test
What is the test?
You only get damages that arise in the usual course of things,
or damages that may reasonably be supposed to be within the contemplation of the parties at the time of contracting.
This is sometimes described as a two-prong test:
(1) Always get damages arising in the usual course of things (called "general damages"); but
(2) only get unusual (called "special") damages if you brought them to the attention of the other party at the time of contracting.
This sounds like one rule for general damages and another, separate rule for special damages. OK to remember it this way, but we will see that the two components boil down to
a single rule: you only get damages if they are "reasonably foreseeable". The "Restatement" gives the foreseeability version of the rule: only get damages if they were reasonably foreseeable as a probable result of the breach.
UCC 2-715(2) uses very similar language.
Why is the foreseeability test the same test as the two-pronged test? Because "damages arising in the usual course of things"--general damages--are always foreseeable; special damages are not normally foreseeable, so you only get those if you have warned the breacher in advance about them (thus making them foreseeable).
Defining foreseeability
How do we tell if something is reasonably foreseeable?
In Hadley: Wasn't it foreseeable to the shipper that the mill would be shut down? The mill owners apparently told the clerk that when the contract was made. So why did the opinion say that the damages were unforeseeable?
No way to tell from the opinion, but it has to do with the rules governing the clerk's authority in 1854; wasn't clear that informing the clerk was the same as informing the firm. This certainly makes it odd that Hadley is such an important precedent since today most would hold that the damages were foreseeable. But this just means that we can't get much help from Hadley in figuring out what foreseeability is.
Other later cases: What about other cases? The other cases give us lots of different verbal formulations: losses have to be--
"a serious possibility", "liable to result", "a real danger", "on the cards."
Does this help? No--we are (as Lord Denning once remarked) swimming on a sea of semantics. Foreseeability has something to do with how likely the damages are, but the court is not ever going to announce some exact probability. How could it? We almost never know the probabilities anyway. It is misplaced precision to look for an exact probability here.
Policies behind foreseeability
What policies guide the decisions about foreseeability? The black letter law provides little guidence (no surprise!), but maybe there are underlying policies that make sense of the decisions.
Unfair surprise
We will start with an obvious policy, one that is often mentioned by the courts. The policy: it is unfair to hold a person liable for losses when a person had (through no fault of that person) no knowledge that losses would be so large. The idea here is to avoid unfair surprise, lack of notice that leads to an unfair imposition of damages.
This is not implausible; fair notice is an important legal concept--law is supposed to be publicly knowable. Think back to what we said last semester about the virtues of having explicit formal rules--fairness,
predictability.
But: How much of a surprise does the liability have to be? aIn most cases, the breacher doesn't know what the damages will be; doesn't know that they will be exactly $48,291.35. Usually, the breach only knows that the damages are likely to fall within a certain range. So the "avoid unfair surprise" idea is that the breacher is not liable for the full amount of damages if they fall way beyond the high end of that range. But this does't not tell us any more than the black letter law; it just is another formulation of the foreseeability test, but it does not give us any better idea about what counts as reasonably foreseeable.
Martians
Beside being uninformative, the "avoid unfair surpise" policy has a serious problem with it. Suppose the Hadley brothers had a spare shaft which they installed when the first shaft broke. So it looks like there will be no damages from a late delivery of the first shaft. But then the Martians land and steal the second shaft (thinking that the shaft is earth's leader). So there are lost profits from the shut-down of the mill?
Are these losses unexpected? Yes--the Martians no doubt did not give advance warning to the shipping company. So, under the "avoid unfair surpise" policy, we cannot. But equally we should not allow the loss to be borne by the Hadley brothers either; they certainly are on no better terms with the Martians than the shippers, and had no advance warning either. So, under the "avoid unfair surpise" policy, we cannot let the loss rest there either.
So why not make the Martians pay? Well some collection problems, but more realistically, suppose it is not the Martians but an earthquake. We can't make the earthquake pay. So we can't put the loss on anybody. But somebody has to bear the loss. The problem is that the theory doesn't tell us who.
Still a lot of courts are influenced by the unfair surprise policy. Attorney argues: "My client did not have notice, so unfair to put the loss on him." Court is sometimes persuaded by this argument. But all this means is that the other attorney is doing a bad job; the other atty should point out that the argument works for his/her client as well: as far as surprise goes, there is no difference between the two parties.
Relative knowledge
But we can get one good suggestion out of this discussion. Maybe what the courts are doing is something like letting the breacher off when the breacher did not know (and should not have known) about the damages (the kind or extent of damages), and the non-breacher did know. That is, maybe it is not just a question of what the breacher did or did not know, but a question of relative knowledge--a question of what the non-breacher knew and did not tell the breacher. So it is a question of what the one knew and the other did not. Without having surveyed the cases (or seen a survey), my guess still is that most cases in which the non-breacher's expectation damages are limited under the foreseeability test are cases in which the non-breacher knew something about damages that the breacher did not. These cases are not like the Martian example in which neither side had any warning.
One way to approach these issues about shared and unshared information is to turn to the economic arguments about expectation damages, the arguments we considered last semester. Shared information issues arise in an interesting way in this context.
Economic arguments for expectation
What about the economic arguments that support the idea of giving full expectation damages?
Let's start with the efficient breach theory. The claim of this theory is that damage rules should promote efficiency. The theory focuses on this situation: there is a seller who has a contract with a buyer--the first buyer, and then the seller kinds a second buyer who will pay him more for the goods, so much more that the seller can sell to the second buyer, fully compensate the first buyer for the breach of that contract, and still come out ahead.
Of course, this requires that the seller have accurate information about the cost of the breach to the first buyer; otherwise, the seller won't make the right calculation about whether the breach will be efficient. To see this suppose that:
Case A: Damages to the first buyer will be $32,000
Neither the seller nor the buyer know this; they both think the damages will be $16,000; they think this because that is what the damages usually are from this kind of breach.
In this case, the seller will be willing to breach if the second buyer pays him $16,000 + his expected profit from selling to the first buyer + some more. Then the buyer will think that he will come out ahead by selling to the second buyer and compensating the first buyer. But the seller will be wrong since the actual cost of the breach is $32,000. Compare this case:
Case B: Damages to the first buyer are $32,000, and
the buyer knows this although the seller does not. Here the seller will, unless the buyer tells him, think the damages are $16,000 and will make the wrong decision about what is an efficient breach.
From an efficiency point of view, what do we want the first buyer to do here? Tell the seller what his damages will be. That we allow the seller to make the right decision about what breaches are efficient.
So we need a rule that gets the information from the buyer to the seller.
Hadley as a mandatory disclosure rule
This is what the Hadley v. Baxendale doctrine does; it tells the first buyer: if you don't disclose the information about damages, you will only get $16,000, not $32,000.
This rule would of course also apply in case A, where the buyer does not have the information about damages. Is that fair? Can argue that it is. It gives the first buyer an incentive to carefully consider his situation to see if there are any special circumstances that would increase his damages above the normal amount. This means that it is more likely that the parties will have accurate information about damages and this increase efficiency.
Precautions
We can make essentially the same point about taking precautions. To provide an incentive to take cost-effective precautions against a breach, a breacher should be liable for damages that are fully compensatory; the damages due should equal the expected cost of the breach.
We can express this as a version of the Hand formula: taking a precaution is cost- effective if and only if cost of taking precautions (proabability of a breach x cost of a breach) - (probability of a breach after the precautions x cost of breach). Therefore, legal liability for damages should equal the cost of a breach. The point is, if the breacher is misinformed about the cost of a breach, he or she will spend too little or too much on precautions. Therefore, if the non-breacher has information about damages, we want it disclosed to the breacher.
Mass markets distinguished
Does this argument work for mass markets? We can raise this question, not just about the efficient breach theory, but also about the relative information argument that we looked at earlier. Let's start with the relative information argument.
The relative information idea is that if the non-breacher knew something about the possible damages that the breacher did not know, that may be a reason to limit the damages the breacher has to pay. The criticism is that this does not make sense in mass transactions.
Suppose I am Federal Express; I ship a million packages a day; you ship one package with me, an extremely valuable one, and I lose it. I didn't know it was so valuable because you did not tell me. The relative knowledge idea is that this may be a reason to say that I do not have to fully compensate you for the loss. But does this make sense?
If I am Federal Express, what do I care about the value of one package? All I care about are total losses--how much I lose in a year. I will budget for losses as a part of the cost of doing business. So all I care about is the average loss: all I need to know is what percentage of packages get lost, and what percentage contain valuable things, where the loss is very high. I don't need to know exactly which packages contain valuable things. So there is nothing unfair about your not telling me about the value of your package since I don't care about that information anyway.
Efficient precautions in mass markets
The same objection works against the precautions argument. If I am Federal Express, I will not adust my level of precautions for each individual package. I will have an overall quality control system, and the amount I spend on that will be based on what I think my average loss will be. I don't need to know what your particular package is worth.
Same point holds for efficient breach theory--although in reality a mass market efficient breach situation is unlikely to arise. Here is what we would have to have: I am a seller carrying out hundreds of transactions a day, and making many efficient breach decisions in a day. This is very unlikely ever to occur, but if it did, all I would care about is the average cost of a breach. It would make no sense to worry about the cost of each individual breach since only the average cost matters.
General conclusion: in mass markets, there is no efficiency rationale
for a disclosure rule about damages.
Under incentive for taking precautions
In fact, applying a Hadley rule in mass markets may be a bad idea because the Hadley rule may lead Federal Express to take too few precautions.
Assume the average loss is $16,000, but there are bigger losses, some as big as $16,000. What we want Federal Express to do is take precautions based on the average loss of $16,000. Now the point is that they will do this only if they know they will be liable for all losses. This means that they will pay between $0 and $32,000 for each loss, depending on how great the damage is. If they face this much liability, they will figure the average loss to be $16,000, and take precautions accordingly.
Now consider what happens with the Hadley rule. Federal Express escapes the high end liability of $32,000; the Hadley rule limits its liabilty--say to a top end limit of $20,000. So what happens when they figure their average legal liability? It will turn out to be less than the average loss because in figuring the liability Federal Express does not include the numbers over $20,000; suppose their average legal liability is $10,000. So if they set their level of precautions based on their average legal liability, they will take too few precautions given what is in fact their average loss. This is clear from the Hand formula. Compare:
cost of precautions (probability of a loss x $16,000) - (probability after precautions x 16,000)
cost of precautions (probability of a loss x $10,000 - (probability after precautions x $10,000)
The second "cost of precautions" figure will be lower than the first.
This is just a version of last semester's argument that to get the right level of precautions, the breacher must be liable for all losses. So why in mass markets do we ever let the breacher off for less?