Homer Simpson once said, “[Y]ou can come up with statistics to prove anything… Forty percent of all people know that.” It’s also been said that “88.2% of statistics are made up on the spot.” The point is that—unless you are an expert in statistical analysis—it’s usually a good idea to take statistics with a grain of salt. For example, when the Obama Administration’s economists tout studies demonstrating that minimum wage increases promote economic growth, one is right to question their assumptions.
Of course, statistical analysis can be enlightening and helpful in crafting public policy. The key is in being sure that you’re working with solid numbers in the first place, and employing sound analytical principles when extrapolating from raw data. For example, the economists at NFIB’s Research Foundation always do a top-notch job in gathering information and reporting on small business trends, from NFIB’s Small Business Optimum Index to the Research Foundation’s studies documenting the adverse effects of minimum wage hikes on economic grown and small business hiring.
But what role should statistics play in lawsuits?
There are undoubtedly times when statistical analysis and expert testimony can be helpful to a party trying to prove a point in court. But, there are times when we would argue that it’s inappropriate to rely on statistics. And last week NFIB Small Business Legal Center joined with other business groups in urging the Supreme Court to reject the practice of certifying class action lawsuits on the basis of statistical assumptions.
According to NFIB Research Foundation surveys, when small businesses are hit with lawsuits, the cost of settlement on average consumes 10% of the business owner’s salary. But, as we pointed out in our filing in Tyson v. Bouaphakeo, in those relatively rare cases where small businesses are hit with class action lawsuits, the potential liabilities—even the cost of settlement—are catastrophic (far in excess of that 10% figure). So it’s essential for federal courts to strictly enforce rules limiting class action lawsuits wherein a class action representative initiates a lawsuit on behalf of every consumer who may have bought a company’s product, or on behalf of every employee who has ever worked at a business.
The federal rules require that before a court can allow such a lawsuit to proceed, plaintiff’s lawyers must offer sufficient evidence to prove that all of the “class members” were similarly situated. In other words, it must be clear that everyone included in the lawsuit was injured by the alleged wrongful conduct, and injured in the same way. So the question in Bouaphakeo is whether that standard can be satisfied by drawing assumptions from statistical models.
In that case employees alleged that they had not been paid for overtime. And they sought to sue on behalf of themselves and every other employee who might have been short-changed; however, they were unable to point to solid evidence as to the hours that the vast majority of the other employees had worked. Instead, they submitted evidence with regard to only a narrow fraction of the employees, and extrapolated from there. They made statistical assumptions as to the average number of overtime hours worked. But, this practice is highly controversial.
For our part, the NFIB Legal Center argued that it is improper to allow a class action suit to proceed in the absence of solid evidence that all of the class members have been injured in the same way. The problem with making statistical assumptions about how class members are injured is that those assumptions do not bear out in reality. And that’s problematic—if a class action is certified—because it results in injustice, both for the defendant-company and to any truly injured individuals.
As we argued, it’s improper to rely on these statistical models because they would ultimately award damages to some individuals who had not been injured (or greater damages than they ought to receive), while conversely denying proper compensation to individuals who may have suffered a more serious injury than the statistical mean average of the representative class. For all of these reasons we maintain that courts should not certify class action lawsuits in reliance on statistical models. And—once more—given how devastating a lawsuit can be for a small business, let alone a class action suit, it’s imperative to ensure that federal rules limiting class action claims are rigidly enforced.
