There seems to be enough interest in this topic that I will contribute my
$0.019999 to the discussion. The topic is _punitive damages_ and the role
they play in the US civil litigation system.
There are obvious pros and cons to punitive damages:
Pro -- They provide a deterrent to certain types of wrongful behavior.
Con -- They encourage plaintiffs with relatively minor losses to file
huge claims. Lawyers call this "playing the lottery".
We seek a win/win solution that preserves the proper use of punitive
damages while minimizing the improper use.
Before proceeding we must be clear about the distinction between harmful
acts and wrongful acts. This distinction is a well-established principle
of law. Typically when a contract is not fulfilled, it is simply breach of
contract, not fraud. In most cases where some loss occurs, it is an
accident, not a crime. For instance, if somebody hits a deer with a rented
car, he presumably wasn't breaking the law, wasn't being reckless, and
wasn't trying to hit the deer. Stuff happens.
Tricky situations arise when the parties are mismatched -- such as when a
large corporation has allegedly harmed an individual consumer. Various
types of award are possible, including
1) no award at all,
2) actual damages, to compensate this individual plaintiff,
3) class-action damages, to compensate all similar parties, and/or
4) punitive damages.
We now discuss each of these possibilities:
1) No award is the right answer in many cases. If I buy a knife and drop
it on my foot, I have been harmed by the knife, but it's not at all clear
that the manufacturer should be held liable. The manufacturer would argue
that it's primarily my own fault, and even though the knife is _not_
perfectly safe, when I choose to use the knife I am assuming the obvious
risks in exchange for the obvious benefits.
2) Actual damages is the right answer in many cases. Suppose my neighbor
hires a professional tree-trimming company. They make a mistake, and a
tree falls into my yard, wiping out my fence and some backyard
furniture. OK, fine, they weren't breaking any laws. They weren't
intending to attack me with a tree. Such accidents are a rare but
essentially unavoidable consequence of their activities. That's why they
are licensed, bonded, and insured. The insurance premium is part of the
price my neighbor contracted for. The insurance company should buy me a
new fence and some new furniture. In rare cases we might have to go to
court to establish the magnitude of the loss that must be remedied.
3) At first glance, it might seem that class-action lawsuits don't change
the story very much, because to a superficial approximation a large number
of small suits has the same effect as a small number of large class-action
suits. On closer examination, however, class actions have two big advantages:
a) One big trial is markedly more efficient than many small trials, and
b) The legal system is very unfair when the parties have mismatched
resources. The big party can hire lots of big-time attorneys who can bury
the little party in paper. This point is highly relevant to the topic of
this note, since one of the arguments sometimes offered in favor of large
punitive damages is that the plaintiff's lawyers are more likely to take
the case, and more likely to mount an effective case, if they think they
might get a share of a large award. Class-action procedures make this less
necessary.
4) We now address the main topic, punitive damages. It is hard to lay down
general rules as to what situations do or do not call for punitive damages,
but here are some thoughts:
a) Suppose some corporation makes money by putting people's lives at
risk. That sounds terrible, doesn't it? But in fact it is perfectly
reasonable. It happens all the time, for the best of reasons. Consider
transportation in particular. A vehicle with every possible safety feature
would be unaffordable and undriveable. So every car ever made contains
tradeoffs involving price, performance, and safety.
There are good tradeoffs and bad tradeoffs. Every year, a number of people
are killed when their cars go off a cliff. The manufacturers know
this. Hmmm, should cars come with parachutes, so the occupants can bail
out in an emergency? Are the manufacturers negligent for not providing
parachutes? I don't think so, because the cost of parachutes is high and
the benefit, on average, is low.
It is ghoulish to make decisions that will cost lives, but it has to be done.
Perhaps the more important issues are
b) Mutually-beneficial versus exploitive tradeoffs, and
c) Sneakiness.
These concepts are related. Let me explain:
Suppose I go to ebay.com and buy some second-hand tools. This is
presumably a mutually beneficial transaction; the tools are worth more to
me than they are to the seller.
In contrast, suppose I go to the store and buy a tool with a _hidden
defect_. This is not a mutually beneficial transaction; I would never
have bought the tool if I'd known about the defect (or the likelihood of
defects). There is also an element of sneakiness; the manufacturer knew
about the defect (or the likelihood of defects) and didn't pass on that
information.
The McDonald's coffee case arguably falls into this category: The store
knew they were selling a product that wasn't what the customer really
wanted, they passed up the opportunity to reduce the risk, and the customer
apparently had no good way of knowing the magnitude of the risk.
In my opinion, any list of exploitive sneaky tricks must put Nestle at or
near the top. These are the folks who in the '70s and '80s sent
representatives dressed up in white "nurse" and "doctor" uniforms to give
out free samples of infant formula to new mothers in poor villages in
Africa. The free samples were just enough to ensure that the mother's milk
would dry up, forcing her to buy formula thereafter. UNICEF estimates that
millions of babies died as a result.
And then there's the tobacco industry. They testify under oath that their
product is not addictive, while they are secretly breeding high-nicotine
strains of tobacco, and giving out free samples on inner-city street corners.
Some exploitations are so egregious that they demand sanctions. Sometimes
criminal trials and prison sentences are appropriate. Sometimes civil
trials and punitive damage awards are appropriate.
Another argument for punitive damages is related to the foregoing point
about sneakiness, and focuses on
d) The probability of getting caught.
Suppose for example a store gets caught overcharging a customer, perhaps by
putting one price on the shelf and putting a higher price in the checkout
stand's barcode lookup database. This happens a lot. If actual damages
(i.e. simple restitution) is the only sanction when they get caught, the
store operators have little incentive to clean up their act, and indeed
have an incentive to make more "mistakes", because they make money whenever
they don't get caught.
In cases like this, it is appropriate to impose a penalty well beyond
simple restitution. The idea is to make sure that dirty tricks don't pay,
even when the probability of detection is less than 100%.
==============
Now we come to the deepest part of the discussion. Suppose it is
appropriate to assess $xxx in punitive damages against the defendant. That
does _NOT_ mean that it is appropriate to pay $xxx to the plaintiff!
In fact, doing so tends to create the appearance of impropriety. It stands
in contrast to the criminal justice system, where the district attorney
(not the victim!) prosecutes the case. The DA has nothing personally at stake.
In physicist's language, many civil damage awards fail the test of
dimensional analysis. In the case where an individual sues a corporation.
the jury wants to assess damages that are large enough to teach the
corporation a lesson. But the resulting sum is completely out of scale for
the individual plaintiff. We are comparing an extensive quantity (the
corporation and all its customers) with an intensive quantity (the
individual plaintiff).
Civil penalties beyond restitution create the appearance of impropriety
even when government officials (not private citizens) are involved. The
cops find a single marijuana seed in a sports car. They make a "civil"
seizure of the car. Did they do so simply in order to make the streets
safer, or because they coveted the car?
(And just BTW, if you try to get your car back after such a seizure, you
will discover there is no presumption of innocence in "civil" procedures.)
So ... finally ... here is a modest proposal:
Anybody should be allowed to sue for punitive damages, but
the punitive damages should never be paid to the plaintiff!
Punitive damages could be paid to a bona-fide charity of the plaintiff's
choosing. This would preserve 100% of the deterrent effect against the
defendant. It would allow the plaintiff to ask for damages in the name of
righteousness rather than selfishness.
Of course one can always sue for actual damages, and for class-action
damages to be paid to the class.
The same rule should apply to the government. The government should not be
allowed to make money by seizing things. Fines and seizures should go to a
victims' trust fund, or to an independent charity.
To summarize: even in cases where vengeance is called for, there should be
No Selfish Vengeance!