May 2004

Did Workers Pay for the Expansion of Products Liability Law?

The expansion of products liability law, which has taken place since the 1960s, ostensibly helps workers. Previously barred by workers' compensation laws from bringing suits against manufacturers of defective products, workers can now sue for recovery from job injuries caused by these products. Research by Alexander (Sasha) Volokh of Harvard University indicates, however, that workers pay for this option in the form of reduced wages. Volokh's research highlights the need to consider the impact of liability system reforms on the entire economy before implementing reforms. Panelists at this May 26 AEI event considered the methodology and implications of Volokh's work.

Alexander "Sasha" Volokh
Harvard University

This paper is an investigation into the effects of the substantial expansion of state products liability law in the 1960s and 1970s on the wages of people who make relevant products and of people who use them on the job. We would expect two separate effects as a result of the expansion of products liability: In an industry that produces dangerous products, when products liability expands an employer becomes less productive-wages might fall, or employment figures might drop. The other effect is on the labor supply side-pre-liability expansion, employees were reimbursed through workers' compensation. Now, thanks to the expansion of products liability, they can sue a distant manufacturer, which is in some sense an extra perk of the job. That leads to another theoretical drop in wages.

Products liability is not a unitary animal, and my research has focused on five doctrines: strict liability, consumer awareness of a product defect, comparative negligence, punitive damages, and inadequate warning as a design defect. The overall effect of these doctrines is to increase both the likelihood that a plaintiff will prevail at trial and the amount of money that he will collect. But how much do workers pay to participate in this lottery? Scholars have established that workers paid for the adoption of workers' comp laws; in fact, wages dropped by almost exactly the amount that a worker might expect to collect from the system. Roughly comparable amounts of money pass through the workers' comp system and the products liability portion of the tort system, so a wage drop of similar magnitude for products liability could be expected.

Regression results indicate that introduction of the doctrine of comparative negligence cases a decrease in wages of 4.3 percent. As a rule, comparative negligence appears to magnify the effects of preexisting doctrines and may be correlated with doctrines that were not included in the analysis. Strict liability comes in with a slightly positive effect (likely due to its replacement of a negligence system that was riddled with exceptions), but the overall effect is basically zero. In a closer look at the manufacturing industry sector, we see a drop of 6.1 percent, compared with 3.3 percent in non-manufacturing industries. It turns out that the supply side effect is about 40 percent of the total effect for comparative negligence, and about 80 percent of the total effect for strict liability.

John R. Lott Jr.
AEI

Firms could have provided a products liability-type insurance, but for some reason they did not. Now we have to ask why product manufacturers were not working this out before the courts got involved. Volokh identifies the changes in laws (through court cases) in specific years, and we see the effects. He talks about the error in measuring the event changes, but it seems to me that something more systematic is going on here. It is costly to wait until a decision has been handed down; once the suit has been filed, the employer has some probability of losing, and workers have some probability of winning. One would think that wages would start changing with the first real shift in legal signals. Even a change of law in another state could lead a reasonable employer or manufacturer to conclude that it is only a matter of time in his jurisdiction.

The demand-side versus supply-side effects are very interesting, but it is not clear what the manufacturing dummy really picks up. It could pick up changes for the desires for the work in certain industries over time. It could pick up changes in comparative advantage in the United States to produce certain types of products over time. More clarity in the definition might make the results more compelling. That leads to some other issues; this analysis is based on a lot of data, and more could be done with it. A simple dummy may be misleading in the case of consistently declining wages over time, for example--the later average will be lower, but it may have nothing to do with the legal shift for which the dummy stands. A breakdown of the data showing time trends and geographic variations would also be great. Finally, the data sets used have a lot of control factors for wages--it would be possible to break down the employment data (by status, type, household information, etc.) to show trends.

Martin Grace
Georgia State University

Somebody has to pay for tort costs. A cursory examination of the scholarship reveals the wide range of transactions influenced by the price of liability insurance: the price of a token on the Philadelphia subway, the price of a ladder, stock prices. The organization and structure of the liability insurance industry are affected by liability rules, as are research and development, innovation, frequency and severity of losses, workers' compensation and wages, and physicians' decisions to leave or enter a market.  Unfortunately, scholars have tended to treat these areas individually, focusing on a one-industry, one-market approach, but the future of liability analysis lies in the examination of the general equilibrium effects of liability expansion or liability reform. Volokh's paper is a first step in that direction.

The classic example of the equilibrium effect involves a rural hospital. With high malpractice premiums applying to rural areas, hospitals can afford neither the premiums nor the risks, and they respond by cutting services. Doctors leave, the area fails to attract new employers because the employers demand medical services, and rural income decreases. There are effects on current local business (there may not be any new ones) and on the local tax base, leaving the person least able to bear risk holding the bag.

If damages are a random event they act as a tax rather than a deterrent and do little or nothing to increase safety. In that case, the liability tax is passed on to the customers, the investors, or the workers. If you look at the cost, we have some 46 percent of each dollar of tort cost going to the victim. The rest of it goes to attorneys' fees and administration. But for workers' compensation, the administrative costs are only 22 percent. That makes the tort system seem very expensive, and those excess costs are borne by someone.

AEI research assistant Kate Rick prepared this summary.