I have some friends in the legal academy who are really enthusiastic about autonomous (self-driving) vehicles. They foresee a future in which these devices carry us to and fro more safely and efficiently than current, human-directed vehicles do.
But many of these same people worry that this future will never come to pass. One concern is that the threat of tort liability down the proverbial road will dissuade possible innovators from making the necessary investments in these devices. And so, some of these advocates have recommended that Congress preempt tort claims against autonomous-driving technologies to encourage their development and diffusion.
In making this argument, boosters sometimes point to the example set by the partial preemption of personal-injury claims associated with the administration of childhood vaccines. As some of you know, in 1986 Congress passed a law that shifted these claims from “traditional” courts and juries to a special forum, which can award damages (funded by an excise tax on the vaccines) upon proof that a claimant’s injury was associated with a covered vaccine. Significantly, this scheme caps pain and suffering damages at $250,000, a ceiling that offers cost certainty to vaccine manufacturers. This statute is today widely regarded as having rescued the childhood vaccine industry from possible ruin, and some people who want to encourage the development of autonomous cars argue that a similar, liability-limiting system (or a more robust preemption scheme) should apply to these vehicles.
Personally, I am more or less agnostic on preemption as it applies to autonomous vehicles. But I am not certain that the vaccine analogy is an especially compelling one. Let me explain why.
Let me start with the circumstances that I regard as creating a paradigm case for preemption premised on a bare threat of ruinous liability (as opposed to, say, preemption designed to ensure the supremacy of federal law, or to avoid interjurisdictional conflicts). Such an argument seems to make the most sense as applied to a device or technology that’s reasonably established as being (1) irreducibly unsafe in some respect; but (2) health- or safety-enhancing, relative to all practical substitutes; and (3) as to which the costs associated with tort judgments are so great as to render continued development or production impractical or undesirable.
The first of these points suggests that tort liability will not encourage further refinements in design, warning, or manufacturing, because it can’t. The second underscores that the product is a socially desirable one, worthy of protection notwithstanding its flaws. The third substantiates an intuition that the product is going to disappear unless the normal mechanics of tort law are somehow constrained.
The history of childhood vaccines, and the status of the vaccination market in the early 1980s, supplies a much better argument for preemption of vaccine lawsuits—along all three dimensions discussed above—than presently exists for preemption of lawsuits involving autonomous cars. In the interest of time, I am going to focus here on the third of these characteristics — the threat to the industry posed by the “normal” application of tort law.
Between 1980 and 1985, 299 lawsuits were filed against the producers of childhood vaccines for injuries allegedly associated with the administration of these vaccines. These lawsuits targeted an industry that was neither especially large nor especially profitable. According to one government report, total sales of childhood vaccines amounted to just $385.8 million for the four-year period between 1981 and 1984, a figured dwarfed by the $3.5 billion in damages demands made in the aforementioned lawsuits. One contemporary media report quoted an employee of a vaccine manufacturer as saying that the dollars demanded in lawsuits pending against the company relating to alleged injury from the pertussis vaccine were 200 times the total sales of the vaccine it produced in 1983. Furthermore, while the precise profitability of the vaccine market at the time is unclear, several of the largest customers had significant buying power, a fact that normally keeps margins relatively low.
A spike in outlays occasioned by product liability suits–through direct damages payments or insurance costs (if insurance were available at all)–thus threatened to drive what was at most a modestly profitable business into the red. The difference between the modest revenues and possibly immense liability exposure also made it difficult for manufacturers to price their vaccines so that they could pass costs on to consumers (especially without being accused of price gouging, a charge that might have negative implications for their other products). Nevertheless, vaccine prices did spike in the early 1980s (rising between 50 and 900 percent during this span, depending upon the vaccine and market), and liability costs received much of the blame.
Predictably, given these circumstances, childhood vaccine manufacturers fled from the business. Per one Congressional report that accompanied the preemption law:
Between 1966 and 1977, half of the commercial corporations producing vaccines in the U.S. ceased production and distribution of these products. Since 1977 this contraction has continued. In 1985, there were only four commercial firms producing and distributing the primary vaccines used in childhood immunization programs. It has been suggested that there are certain characteristics of the vaccine market that make vaccine production and distribution relatively unattractive from a commercial perspective, which in turn has acted to discourage producers from entering or staying in the business of producing vaccines.
The four remaining domestic, private participants could very credibly threaten to leave the business themselves. Most of these entities had many other product lines to keep their corporate coffers full, if they chose to leave vaccine manufacture behind:
- Other products marketed by Wyeth Laboratories’ parent (American Home Products Corporation) included Anacin, Dristan, Preparation H, Crunch ‘n’ Munch, Woolite, Pam Cooking Spray, and Easy-Off oven cleaner.
- Lederle Laboratories’ parent (American Cyanamid Company) sold products including Lady’s Choice antiperspirant, Old Spice deodorant, Breck shampoo, Formica brand laminates, and Pine-Sol liquid cleaner.
- A third vaccine producer, Merck, claimed a robust portfolio of profitable prescription drugs.
The other drugs and consumer goods sold by these companies were big business, compared to vaccines: according to their respective annual reports, American Home Products, American Cyanamid, and Merck boasted 1983 revenues of $4,857 million, $3,536 million, and $3,246 million, respectively.
Meanwhile, no guarantee existed that new entrants would replace these dropouts. Development of a new vaccine was then, as it is now, a costly affair. Other companies conceivably might purchase an existing vaccine business from an existing participant—but why would they, given the market dynamics discussed above?
* * *
In sum, the threat to the vaccine market in the mid-1980s was real, not just theoretical. Manufacturers’ threat of exit was bolstered by the facts that other companies had abandoned the business, childhood vaccines were a small, relatively flat, and not especially lucrative business, and these companies had other product irons in the fire. Meanwhile, substantial barriers blocked new entrants.
For the autonomous vehicles / vaccines analogy to prove persuasive, I’m going to need to see similar evidence that tort liability poses a real, as opposed to an existential threat to the development of autonomous cars. (I’ll also need to hear more about how autonomous vehicles will in fact be safer than human-operated vehicles, but that’s another story.)