Econ 522 Economics of Law Dan Quint Fall 2015 Lecture 16

Last week, we began tort law How to create incentives to avoid accidental harms? Strict liability and negligence rules Harm, causation, breach of duty Simple economic model unilateral harm reduce behavior to a one-dimensional choice for each player, how careful to be (“precaution”) think about effect of liability rule on precaution and on activity levels of both players

Model of unilateral harm x level of precaution w marginal cost of precaution p(x) probability of an accident A cost of an accident Model of unilateral harm efficient precaution: minx { wx + p(x) A } $ wx + p(x) A (Total Social Cost) Let’s graph some important quantities – with x (the level of precaution taken) on the x axis, and money on the y We said the probability of an accident, p(x), is decreasing in the level of precaution Since A is a constant, this means p(x) A – the expected cost of accidents – is downward-sloping The cost of precaution is just a line, w times x And we can add these together to calculate the total expected social cost related to accidents, wx + p(x) A And now our goal in tort law becomes clear: to achieve the efficient level of precaution, which is the level of x that minimizes the total social cost So we’ll put this in gold, and call it x* (Of course, this assumes precaution is continuous – that you can take any amount you choose Some precaution is discrete – you either turn on your headlights while you’re driving, or you don’t But the conceptual goal is the same: to get headlight use when the reduction in accidents is greater than the cost And to not get headlight use when the cost would be greater than the reduction in accidents) wx (Cost of Precaution) p(x) A (Cost of Accidents) x* (Efficient Level of Precaution) Precaution (x)

Benchmark: what happens if injurer faces no liability at all? Injurer doesn’t perceive accidents as part of private cost No incentive to take any costly precaution No incentive to reduce activity, which will be above efficient level Victim bears cost of accidents as part of private cost Victim’s private cost = social cost Faces incentive to take efficient amount of precaution Faces incentive to reduce activity to efficient level A rule of no liability leads to… no injurer precaution and overly high injurer activity but efficient victim precaution and victim activity level

Benchmark: No Liability Injurer Precaution Victim Precaution Injurer Activity Victim Activity No Liability Zero Efficient Too High Efficient

Another way to think about this – with no liability… Injurer precaution Social cost: cost of precaution Social benefit: reduction in accidents Private cost to injurer: cost of precaution Private benefit: nothing! PC = SC, but PB < SB, so injurer takes less than efficient precaution Victim precaution Social cost: cost of precaution Social benefit: reduction in accidents Private cost to victim: cost of precaution Private benefit: reduction in accidents PC = SC and PB = SB, so victim takes efficient precaution

Another way to think about this – with no liability… Injurer activity Social cost: cost of activity, plus increase in accidents Social benefit: benefit of activity Private cost to injurer: cost of activity only Private benefit: benefit of activity PB = SB, but PC < SC, so injurer takes higher than efficient activity level Victim activity Social cost: cost of activity, plus increase in accidents Social benefit: benefit of activity Private cost to victim: cost of activity, plus increase in accidents Private benefit: benefit of activity PC = SC and PB = SB, so victim takes efficient activity level

Next: what happens under a strict liability rule? 7 7

Strict Liability Perfect compensation: damages D = A Under strict liability… Injurer pays damages for any accidents he causes So injurer bears cost of accidents, plus his own precaution Injurer internalizes externality his actions cause  chooses efficiently Victim is fully insured, no incentive for precaution Next, suppose we are in a world with strict liability, where in the event of an accident, the injurer will have to pay damages D to the victim And suppose that compensation is perfect – D = A, so damages exactly compensate for the harm. Go back to the example of reflective clothing Now the bicyclist knows that if he’s in an accident, he’ll incur harm of A and receive damages of D, so now the costs he faces are w x + p(x) (A – D) = wx So the cyclist minimizes this by setting x = 0 Under strict liability, the victim doesn’t bear the cost of accidents, so he has no reason to take any precaution Is this efficient? No. The total social cost of accidents is wx + p(x) A the transfer of money from injurer to victim is just a transfer So it would be efficient to set a higher level of precaution. So a strict liability rule leads the victim to select an inefficiently low level of precaution. But on the other hand, if it’s the injurer who affects the probability of accidents, strict liability ends up working pretty well The injurer knows if there’s an accident, he’ll have to pay damages of D, so he minimizes wx + p(x) D and with D = A, this is wx + p(x) A So he’s facing the same problem as minimizing the total social cost of accidents – so he chooses x efficiently! (This is because, under strict liability, the victim is actually indifferent between accident and no accident – so the injurer’s actions cause no externality, so he chooses efficiently.) So a strict liability rule leads the injurer to take efficient precaution. 8 8

Strict Liability (Damages = A) Injurer’s private cost is p(x) A + wx Minimized at efficient precaution level x = x* Victim’s private cost is just wx Minimized at x = 0 So rule of strict liability leads to efficient precaution by injurers, no precaution by victims $ Private cost to injurer Private cost to victim wx + p(x) A wx Next, we consider Rule 2: Strict Liability Suppose we are in a world with strict liability, where in the event of an accident, the injurer will have to pay damages D to the victim And suppose that compensation is perfect – D = A, so the damages exactly compensate the victim for the harm. Go back to the example of reflective clothing Now the bicyclist knows that if he’s in an accident, he’ll incur harm of A and receive damages of D, so now the costs he faces are w x + p(x) (A – D) = wx So the cyclist minimizes this by setting x = 0 Under strict liability, the victim doesn’t bear the cost of accidents, so he has no reason to take any precaution Is this efficient? No. The total social cost of accidents is wx + p(x) A the transfer of money from injurer to victim is just a transfer So it would be efficient to set a higher level of precaution. So a strict liability rule leads the victim to select an inefficiently low level of precaution. But on the other hand, if it’s the injurer who affects the probability of accidents, strict liability ends up working pretty well The injurer knows if there’s an accident, he’ll have to pay damages of D, so he minimizes wx + p(x) D and with D = A, this is So he’s facing the same problem as minimizing the total social cost of accidents – so he chooses x efficiently! (This is because, under strict liability, the victim is actually indifferent between accident and no accident – so the injurer’s actions cause no externality, so he chooses efficiently.) So a strict liability rule leads the injurer to take efficient precaution. p(x) A x* x 9 9

Strict Liability Injurer Precaution Victim Precaution Injurer Activity Victim Activity No Liability Zero Efficient Too High Efficient Strict Liability Efficient Zero 10 10

What about activity level? Under strict liability, injurer internalizes cost of accidents Weighs benefit from driving against cost of accidents Takes efficient activity level Under strict liability, victim does not bear cost of accidents Ignores cost of accidents when deciding how much to bike Sets inefficiently high activity level A rule of strict liability leads to the efficiently level of injurer activity, but an inefficiently high level of victim activity The exact opposite will happen under strict liability Under strict liability, you know you will be compensated for any accidents, so you don’t worry about their costs Therefore, you take insufficient precautions, and you set too high an activity level On the other hand, I know I’m paying for any accidents that happen, so I consider their costs when I make my decisions I take efficient precautions and I set an efficient activity level. 11 11

Strict Liability Injurer Precaution Victim Precaution Injurer Activity Victim Activity No Liability Zero Efficient Too High Efficient Strict Liability Efficient Zero Efficient Too High 12 12

So… For both precaution and activity level… “No liability” leads to inefficient behavior by injurer, efficient behavior by victim “Strict liability” leads to efficient behavior by injurer, inefficient behavior by victim Reminiscent of paradox of compensation One rule sets multiple incentives… …can’t get them all right But in tort law, we have a trick… Our results so far suggest that when only the victim is able to take precautions to prevent the accident, a rule of no liability is preferable And when only the injurer is able to take precautions, a rule of strict liability is preferable. (This is exactly analogous to the problem we saw in contract law, of allocating a risk to the lowest-cost avoider When one party – in this case, victim or injurer – can take precautions and the other cannot, that first party is the low-cost avoider Making them bear the cost of the accident leads to efficient unilateral precaution.) With contracts, we saw that there was no breach remedy that set efficient incentives for both the promisor and the promisee – the paradox of compensation Here, we find the same thing: a rule that sets the correct incentive for injurer precaution leads to too little victim precaution, and vice versa So in settings with bilateral precaution – situations where both parties have the possibility of precaution – neither of these rules will lead to efficiency (With bilateral precaution, things get a little more complicated – you’ll see an example of this on the homework.) However, in tort law, there is a trick We already defined a negligence rule, under which the injurer is liable for harm only if he did not take certain precautions That is, in tort law, we can make damages depend not only on whether an accident occurred, but also on the level of precaution that the parties took ahead of time This may allow us to get both incentives right. 13 13

Negligence

Negligence Rule Negligence rule: injurer is only liable if he breached the duty of due care Within our model: Legal standard of care xn Injurer owes damages if precaution level was below that level x < xn  D = A x ³ xn  D = 0 So on our graph from before, private cost to injurer is… wx + p(x) A for x < xn wx for x ³ xn Best shot at achieving efficiency is to set xn = x* Rule 3: Negligence Rule Let’s think first about just injurer precaution A negligence rule says that the injurer is liable for damages if his precaution level was below the legal standard of care, xn So for x < xn, damages would be D = A For x >= xn, damages would be D = 0 If we go back to the graphs we drew earlier, then, the total expected cost to the injurer (from both precaution and expected damages) will be wx + p(x) A for x < xn, and just wx for x >= xn, since when x >= xn, the injurer is not liable for damages. How this actually looks will depend on how xn, the legal standard for precaution, relates to x*, the efficient level of precaution For now, let’s assume they’re equal, that is, xn = x* (Our goal is to design an efficient legal system; so for now, let’s assume the court can set the efficient legal standard for negligence. We’ll see later what happens when it can’t.)

“Simple Negligence”: Injurer precaution $ Private cost to injurer wx + p(x) A wx What will injurer do? At x < x*, we know that wx + p(x)A is decreasing; and so the injurer’s expected costs are decreasing Similarly, we know that wx is increasing, so above x*, the injurer’s expected costs, which are just wx, are increasing So expected costs are minimized by setting x = x*, the efficient level. So a negligence rule, with a legal standard of care x~ equal to the efficient level x*, leads to an efficient level of injurer precaution. But now, what about the victim’s incentives? The victim knows that under a negligence rule, the injurer will take precaution x*; and therefore, if an accident occurs, the injurer will not be liable So now the victim is facing the full cost of any accident which occurs; which means that the victim also takes the efficient level of precaution. So now we’ve found a solution to the paradox of compensation A negligence rule, combined with an efficient legal standard for care, leads to efficient precaution by both injurer and victim. Which is pretty cool. (We’ll come back to what happens when standard for care is set incorrectly later.) p(x) A xn = x* x Private cost is wx + p(x) A if x < xn, only wx otherwise If standard of care is set efficiently (xn = x*), injurer minimizes private cost by taking efficient precaution

“Simple Negligence”: Victim precaution What about victim? We just said, injurer will take efficient precaution Which means injurer will not be liable So victim bears costs of any accidents (Victim bears residual risk) So victim’s private cost is wx + p(x) A Victim minimizes private cost by taking efficient level of precaution too! $ Private cost to victim (assuming injurer takes efficient level of precaution and is therefore not liable for damages) wx + p(x) A wx p(x) A x*

Simple Negligence Injurer Precaution Victim Precaution Injurer Activity Victim Activity No Liability Zero Efficient Too High Efficient Strict Liability Efficient Zero Efficient Too High Simple Negligence, xn = x* Efficient Efficient

Other negligence rules Rule we just saw is called “simple negligence” Only consider injurer’s actions in determining liability But we could also consider whether victim was negligence in deciding whether injurer is liable “Negligence with a defense of contributory negligence” – injurer owes nothing if victim was also negligent “Comparative negligence” – if both were negligent, share cost “Strict liability with defense of contributory negligence” – injurer is liable (even if he wasn’t negligence), unless victim was negligent Any of these rules (with efficient standard of care) will lead to efficient precaution by both parties! (We could show this for each the way we already did for simple negligence, but that would get boring and repetitive, so instead, let’s consider a different example, with discrete precaution)

Discrete example of bilateral precaution A $1,000 w $20 for either party p 10% / 6% / 2% No “levels” of precaution – each party can either take precaution or not Each accident causes $1,000 of harm Precaution costs $20 for each party Chance of an accident is 10% if nobody takes precaution 6% if one party takes precaution 2% if both parties take precaution Note that precaution is efficient for both parties Costs $20; reduces expected accidents by 4% X $1,000 = $40

Negligence with a Defense of Contributory Negligence A $1,000 w $20 for either party p 10% / 6% / 2% Injurer is liable if he failed to take precaution… Unless victim did too Precaution is always best-response for victim If injurer is not taking precaution, victim wants to avoid liability If injurer is taking precaution, victim bears residual risk, wants to minimize accidents For injurer, precaution is the best-response to precaution “Both take precaution” is the only Nash equilibrium And, is the efficient outcome victim pays for precaution, and any accidents that happen injurer pays for precaution, not liable for accidents Victim Precaution None -20, -40 -20, -60 Precaution Injurer -60, -20 0, -100 None 21 21

Comparative Negligence A $1,000 w $20 for either party p 10% / 6% / 2% If both parties were negligent… …divide cost proportionally Precaution is again always a best response for victim Now it’s a best response for injurer too Again, “both take precaution” is the only equilibrium (And the efficient outcome) Victim Precaution None -20, -40 -20, -60 Precaution Injurer -60, -20 -50, -50 None 22 22

Strict Liability with a Defense of Contributory Negligence A $1,000 w $20 for either party p 10% / 6% / 2% Now, injurer is liable, regardless of whether he took precaution… …unless victim was negligent Once again, “both take precaution” is the only equilibrium Victim Precaution None -40, -20 -20, -60 Precaution Injurer -60, -20 0, -100 None 23 23

Negligence Injurer Precaution Victim Precaution Injurer Activity Victim Activity No Liability Zero Efficient Too High Efficient Strict Liability Efficient Zero Efficient Too High Simple Negligence Efficient Efficient assuming all relevant standards of care are set to the efficient levels Negligence with a Defense of Contributory Negligence Efficient Efficient (We can also think about this via externalities. Under strict liability, you are fully compensated for any accidents I cause So my choice of how much to drive doesn’t affect your well-being So there is no externality, and I drive the efficient amount. On the other hand, under strict liability, if you bicycle more, it increases the likelihood that I hit you and have to pay, so it imposes a negative externality on me, so you do it too much.) Comparative Negligence Efficient Efficient Strict Liability with Defense of Contributory Negligence Efficient Efficient

What about activity levels under a negligence rule? Simple negligence: injurer is only liable if he was negligent Leads injurer to take efficient precaution, so injurer expects to not be liable for any accidents that do occur So injurer ignores cost of accidents when deciding on activity level (how much to drive) Injurer drives carefully, but still drives too much Victim bears “residual risk” Victim bikes carefully, and bikes efficient amount

Adding activity levels to our results on precaution… Injurer Precaution Victim Precaution Injurer Activity Victim Activity No Liability Zero Efficient Too High Efficient Strict Liability Efficient Zero Efficient Too High Simple Negligence Efficient Efficient Too High Efficient Negligence with a Defense of Contributory Negligence Efficient Efficient Comparative Negligence Efficient Efficient Strict Liability with Defense of Contributory Negligence Efficient Efficient

Negligence with Defense of Contributory Negligence, and Comparative Negligence Either rule: efficient precaution by both parties Either rule: if neither party was negligent, injurer does not owe damages So victim is residual risk bearer (pays for accidents) So victim weighs cost of accidents against benefits of activity, takes efficient activity level Injurer ignores cost of accidents, takes inefficiently high activity level

Negligence Injurer Precaution Victim Precaution Injurer Activity Victim Activity No Liability Zero Efficient Too High Efficient Strict Liability Efficient Zero Efficient Too High Simple Negligence Efficient Efficient Too High Efficient Negligence with a Defense of Contributory Negligence Efficient Efficient Too High Efficient Comparative Negligence Efficient Efficient Too High Efficient Strict Liability with Defense of Contributory Negligence Efficient Efficient

Strict Liability with Defense of Contributory Negligence If victim is not negligent, injurer is liable Leads to efficient precaution by both, so injurer is liable Injurer is residual risk bearer Injurer weighs cost of accidents against benefits of activity, takes efficient activity level Victim ignores cost of accidents, takes inefficient high activity level

Negligence Injurer Precaution Victim Precaution Injurer Activity Victim Activity No Liability Zero Efficient Too High Efficient Strict Liability Efficient Zero Efficient Too High Simple Negligence Efficient Efficient Too High Efficient Negligence with a Defense of Contributory Negligence Efficient Efficient Too High Efficient Comparative Negligence Efficient Efficient Too High Efficient Strict Liability with Defense of Contributory Negligence Efficient Efficient Efficient Too High

An easier (perhaps) way to understand all of this take precaution only to AVOID liability precaution is efficient, but activity level is too high Injurer Precaution Victim Precaution Injurer Activity Victim Activity No Liability Zero Efficient Too High Efficient Strict Liability Efficient Zero Efficient Too High Simple Negligence Efficient Efficient Too High Efficient Negligence with a Defense of Contributory Negligence Efficient Efficient Too High Efficient Comparative Negligence Efficient Efficient Too High Efficient Strict Liability with Defense of Contributory Negligence Efficient Efficient Efficient Too High

An easier (perhaps) way to understand all of this to reduce accidents, since he bears their cost precaution and activity level are both efficient Injurer Precaution Victim Precaution Injurer Activity Victim Activity No Liability Zero Efficient Too High Efficient Strict Liability Efficient Zero Efficient Too High Simple Negligence Efficient Efficient Too High Efficient Negligence with a Defense of Contributory Negligence Efficient Efficient Too High Efficient Comparative Negligence Efficient Efficient Too High Efficient Strict Liability with Defense of Contributory Negligence Efficient Efficient Efficient Too High

With each negligence rule… One party can avoid liability by taking efficient precaution Leads to efficient precaution But inefficient activity level Other party is the residual risk bearer – even when he takes precaution, he is still liable And also efficient activity level Who should bear residual risk? One way to answer is to ask whose activity level has greater impact on efficiency

So which rule is best? “Put the incentive where it does the most good” Efficient rule depends on which choices have greatest impact If only injurer’s choices (precaution + activity) matter  strict liability is better rule If bilateral precaution  negligence Which negligence rule – depends whose activity level is more important Friedman (citing Posner): this is why very dangerous activities often covered by strict liability Blasting with dynamite, keeping a lion as a pet Even with proper precaution, still very dangerous, so injurer activity level is important Which rule is the most efficient, then, depends on the situation That is, it depends on whose choices have the bigger impact. In situations with unilateral precaution – situations where only one party needs to take steps to avoid accidents – no liability or strict liability work perfectly well In situation with bilateral precaution, negligence rules give appropriate incentives for precaution on both sides However, under any negligence rule, one of the two parties still bears whatever losses do occur – one party is the residual bearer of the harm of accidents, and the other party is not Any negligence rule yields an efficient activity level by the residual risk bearer, but an inefficient activity level by the other party Thus, the optimal rule depends on whose activity level is most important. Friedman mentions that Posner uses this to explain why highly dangerous activities – blasting with dynamite, or keeping a lion as a pet – are often governed by strict liability Wikipedia offers the example of a tiger rehabilitation center – if a tiger escapes and causes injuries, the owner is responsible, no matter how careful he was if an activity is dangerous enough, the only meaningful type of precaution may be to not do them at all Strict liability leads to efficient levels of both care (if you choose to do them) and a choice of whether or not to do them in the first place.

Friedman: activity is just unobservable precaution Activity is just another type of precaution, but type where court can’t determine efficient level Court can tell inefficient for me to drive at night with headlights off Can’t tell how many miles it’s efficient for me to drive Determination of negligence can only be based on observable precaution, not unobservable Negligence rule leads to efficient levels of observable precaution by both parties Simple negligence leads only to efficient observable precaution by injurer, but efficient precaution by victim as well Strict liability leads to efficient observable and unobservable precaution by injurer, but no precaution by victim Friedman has a different take on activity levels He argues that reducing your activity level is just another type of precaution But it’s a type of precaution where it’s impossible for the court to determine the efficient level (A court might be able to figure out that it’s efficient for me to drive with my headlights on at night, but unable to figure out how many miles it’s optimal for me to drive in a given day Therefore, a court might find me negligent if I was driving without headlights But it’s hard for a court to decide whether a particular trip was socially efficient, and therefore whether I was negligent by being in my car!) Instead of distinguishing precaution from activity levels, Friedman distinguishes observable precaution from unobservable precaution He points out that any negligence rule only causes an efficient level of observable precaution, since that’s all that can be used to determine whether someone exercised due care or was negligent So a negligence rule does not lead to an efficient level of unobservable precaution Strict liability, on the other hand, leads the injurer to internalize the cost of accidents, so it leads to efficient levels of both observable and unobservable precaution – the same result as we already saw, just in different words

What if injurer is a business?

Steven Shavell, Strict Liability Versus Negligence Focuses on injurer precaution and activity Compares strict liability to negligence rules Accidents between strangers (what we’ve been doing): “Under a negligence rule, all that an injurer has to do to avoid the possibility of liability is to make sure to exercise due care if he engages in his activity. Consequently he will not be motivated to consider the effect on accident losses of his choice of whether to engage in his activity or, more generally, of the level at which to engage in his activity; he will choose his level of activity in accordance only with the personal benefits so derived. But surely an increase in his level of activity will typically raise expected accident losses. Thus he will be led to choose too high a level of activity.” The Shavell paper, “Strict Liability versus Negligence,” is about exactly the incentives we’ve been discussing Shavell refers to the types of cases we’ve been considering so far “accidents between strangers” That is, when a car hits a bicyclist, there is no preexisting relationship between the two parties However, Shavell also looks at cases where the injurer is a business, engaged in selling some product (either to the victim or to someone else) Under the assumption of competitive markets, this changes the story significantly. Shavell assumes it is only injurer precaution that affects the likelihood of accidents – that is, he ignores victim precaution He looks at a number of different cases Under injurer precaution, a rule of no liability is obviously bad, so he focus on comparing negligence to strict liability. And, like we’ve been doing, he assumes there is no insurance. First, as we’ve already seen, is the case of accidents between strangers As he puts it, “injurers and victims are strangers, neither are sellers of a product, and injurers may choose to engage in an activity which puts victims at risk” He gives a nice explanation of what happens under a simple negligence rule:

Steven Shavell, Strict Liability Versus Negligence Whereas under strictly liability… “Because an injurer must pay for losses whenever he in involved in an accident, he will be induced to consider the effect on accident losses of both his level of care and his level of activity. His decisions will therefore be efficient. Because drivers will be liable for losses sustained by pedestrians, they will decide not only to exercise due care in driving but also to drive only when the utility gained from it outweights expected liability payments to pedestrians.” (This is exactly what we had already concluded…)

Steven Shavell, Strict Liability Versus Negligence Injurer Precaution Injurer Activity ACCIDENTS BETWEEN STRANGERS Simple Negligence Efficient Too High Strict Liability Efficient Efficient The other cases Shavell considers are when the injurer is a business, rather than a private citizen The incentives for taking precaution as a business owner are very similar to those of a private citizen But there is a new element – price! Shavell considers businesses in a perfectly competitive industry – that is, he assumes competition drives prices down to marginal cost So if taking precaution increases a business’s costs, that cost is passed on to consumers through higher prices

Next case: accidents between “sellers and strangers” Injurer is in a competitive business, but not with victim victim is not injurer’s customer, but a stranger Example: taxi drivers provide service to their passengers risk hitting other pedestrians victims are not their own customers Shavell assumes perfect competition Price = marginal cost of “production” Sales = number of passengers who demand rides at that price Next, Shavell considers accidents between sellers and strangers That is, injurers are engaged in some sort of competitive business, but not with their victims He gives the example of taxi drivers – taxis provide a service to passengers, but still run the risk of hitting other pedestrians who aren’t their passengers Shavell assumes that there is perfect competition – taxis lower their prices to compete against each other, up till prices equal the costs of “production”; and the number of passengers who demand rides at those prices determine the level of sales

Accidents between businesses and strangers Strict liability Taxi drivers pay for accidents, set x = x* to minimize private cost Perfect competition  cost of remaining accidents is built into price Taxi passengers face price that includes cost of accidents Passengers internalize risk of accidents, demand efficient number of rides Negligence rule Taxi drivers still take efficient precaution, to avoid liability But since drivers don’t bear residual risk, cost of accidents not built into price Passengers face prices that are too low (lower than social cost of a taxi ride) Demand for taxi rides inefficiently high For accidents between strangers, strict liability led to efficient precaution and efficient activity For accidents between businesses and strangers, the results are the same, but the reason is different Under strict liability, taxi drivers pay for accidents, so they will take the efficient level of precaution – it’s cheaper for them to take precaution than to pay for more accidents And the expected costs of any remaining accidents is still borne by the taxi drivers So under competition, it is built into the price of a taxi ride (That is, once taxi fares reach the level that just covers costs plus expected damage payments, taxi drivers stop lowering prices, so that sets the price.) This means that taxi passengers face the “socially optimal” price – given efficient precaution, they now internalize the cost of accidents, so they take the efficient number of taxi rides A negligence rule led to efficient precaution but inefficiently high activity for accidents between strangers Here, the result is the again the same, but again, for a different reason Under a negligence rule, taxi drivers take efficient precaution, to avoid liability for any accidents that occur But once they’re taking efficient precaution, they’re no longer liable, so they do not bear the costs of accidents So the cost of accidents is not built into prices So passengers face prices that are too low – taxi passengers don’t internalize the cost of accidents when they decide how often to ride So the demand for taxi rides will be inefficiently high – again, we get an inefficiently high activity level

Steven Shavell, Strict Liability Versus Negligence Injurer Precaution Injurer Activity ACCIDENTS BETWEEN STRANGERS Simple Negligence Efficient Too High Strict Liability Efficient Efficient ACCIDENTS BETWEEN BUSINESSES AND STRANGERS The two cases look exactly the same, but for different reasons Simple Negligence Efficient Too High Strict Liability Efficient Efficient

Final case: accidents between businesses and their own customers Example: restaurants taking precaution to reduce risk of food poisoning How accurately do customers perceive risks? 1. Customers can accurately judge risk of each restaurant 2. Customers can accurately judge average level of risk, but not differences across restaurants 3. Customers ignorant of risks The third case that Shavell considers is that of accidents between sellers and their customers Here, he uses the example of restaurants taking precautions to reduce the risk of food poisoning (He also points out that accidents between sellers and their employees are very similar) In this case, it ends up vitally important how accurately customers perceive risks He looks at three separate cases: customers can accurately perceive the risk of each restaurant customers can accurately perceive the average level of risk, but not differences between different restaurants customers are just generally ignorant of the risks

Accidents between businesses and their own customers: strict liability Seller bears cost of accidents  efficient precaution Seller bears residual risk  expected cost of accidents is built into prices Even if customers don’t perceive risk, price leads them to make efficient choices Price of shellfish = cost of shellfish + expected cost of food poisoning Even if I don’t know that, I buy shellfish when benefit > price, so I’m forced to choose efficiently Under strict liability, customer perception of risk ends up not mattering after all Under strict liability, the seller bears the cost of accidents, so he takes efficient precaution to prevent them And since he still bears the remaining risk of accidents, their expected cost is built into the price of meals That is, menu prices include a premium for expected damage payments when food poisoning does occur Which means that regardless of whether they perceive risk, customers make efficient choices about how often to eat out, because they explicitly see the cost of risk through prices That is, if I like seafood, I don’t have to know that if I eat raw shellfish, I have a 1-in-100 chance of an unpleasant experience that costs me $500 I don’t need to know this because the cost of a raw shellfish meal is $5 more than it would be otherwise, since the restaurant needs to charge this just to break even since if I ever do get sick, he knows I’ll sue So the prices I face include the expected cost of food poisoning; so even if I don’t realize that, I choose efficiently.

Accidents between businesses and their own customers Risk Perception? Seller Precaution Buyer Activity Strict Liability Yes Efficient Efficient No Efficient Efficient

Accidents between businesses and their own customers: negligence Restaurants take efficient precaution, to avoid liability But since they avoid liability, cost of accidents not built into prices If customers perceive risk correctly, no problem Weigh benefit of meal versus price + expected pain due to food poisoning Demand efficient number of meals But if customers don’t perceive risk, they’ll demand inefficiently many dangerous meals Under negligence, however, this doesn’t work Under negligence, restaurants will still take efficient precaution, to avoid liability (they’ll keep the kitchen clean, wash hands after using the bathroom, etc.) But since they then won’t be liable when accidents do occur, prices won’t reflect these risks If customers correctly perceive risks, this won’t matter – customers will consider the money cost of the meal, plus the expected cost due to food poisoning, and will choose efficiently But if customers underestimate the risk of food poisoning, they’ll order an inefficiently large number of risky meals – the level of activity will be too high

Accidents between businesses and their own customers Risk Perception? Seller Precaution Buyer Activity Strict Liability Yes Efficient Efficient No Efficient Efficient Negligence Yes Efficient Efficient No Efficient Too High

Accidents between businesses and their own customers: no liability If customers correctly judge risks… Restaurants take efficient precaution to attract customers And customers demand efficient number of meals If customers can only judge average level of risk… Restaurants take no precautions But customers know this, demand efficient (low) number of meals If customers are oblivious to risk… Cost of food poisoning not built into prices Customers demand inefficiently high number of meals We can also look at what will happen under a rule of no liability If customers correctly judge the risk, then sellers will still take efficient precaution – it gives them a way to attract customers If the restaurant can spend $3 to make a meal $5 cheaper in expected costs, they could charge $4 more for the meal and still get more customers Competition means that this will happen So if risk is judged accurately, sellers will take efficient precautions, and buyers, judging risk correctly, will buy the efficient number of meals Next, suppose customers can accurately judge the average level of risk across all restaurants, but can’t differentiate between restaurants In this case, restaurants have no reason to take precaution – since they won’t be rewarded with higher sales, and they won’t be liable if there are accidents – so precaution will be zero But, at least customers will know that precaution is low and the product is risky, so they’ll eat the efficient number of meals Under no liability in a city without any health inspectors, maybe people don’t eat much sushi Finally, consider the case where customers are just completely oblivious to the risk of the product Like the last case, there is no reason for sellers to take any precaution And since there is no liability, risk will not be built into prices And since customers can’t perceive risk, not only is precaution too low, but activity is too high – customers don’t consider the risks when deciding how often to eat out

Accidents between businesses and their own customers Risk Perception? Seller Precaution Buyer Activity Strict Liability Yes Efficient Efficient No Efficient Efficient Negligence Yes Efficient Efficient No Efficient Too High No Liability Yes Efficient Efficient Average None Efficient No None Too High

Next few lectures… How do we determine legal standard for negligence? What happens if we get it wrong? What happens when the world is more complicated than we’ve been imagining so far?