When drafting its Principles for Aggregated Litigation, the American Law Institute considered a novel idea.  instead of applying the law of each of the fifty states to a multi-state class action based on state law (as the Supreme Court had long held was required by due process), a court could instead apply the "average" state law, determined averaging the probability a defendant would be held liable under each jurisdiction substantive law.

The ALI ultimately rejected the proposal. But Harvard Law professor David Rosenberg, joined by 3L Luke McCloud, thinks that courts should actually adopt the proposal when deciding whether to certify a class.

So what’s their pitch?

The rejection of the average law solution is profoundly mistaken. It stems from the prevailing view among courts and commentators that the nature of the governing law and businesses‟ understanding and response to it at the time of the underlying conduct is the same regardless of whether the contemplated activity involves an intrastate or interstate risk. Our principal contribution is a basic, straightforward point: the average of the differing state laws is in reality the actual law that in fact ultimately governs the choice a business will make and expresses the choice the multiple states involved expect and presumably want the business to make regarding whether and how safely it should engage in activities involving interstate risk.

(Footnotes omitted.)

The authors make an intriguing theoretical argument, but it’s woefully incomplete. Without diving too deep, a few of the questions it leaves unanswered:

Who decides what the "average law" is? Given the novelty of the proposal, a plaintiff who just argues "Well, we’ll use average state law," is going to have to show that the proposal works in practice. And part of that will be showing the court that it is possible to apply average state law in both a fair and efficient way.  Deciding the best "average" will likely require as much work from the court as crafting an "Esperanto" jury instruction.  

Does the authors’ model account for changes in law? As any practicing lawyer can tell you, each new case issued in a state court changes the odds of proving liability, sometimes subtly, sometimes less so. It’s certainly possible for a mathematical model to account for changes in information, but it’s not something the authors account for here.  And accounting for changes will inevitably complicate the model.

What about Shady Grove? This is the doctrinal objection, and it’s a huge one. Because Rule 23 gets its legal force from the Rules Enabling Act, it cannot change the substantive rights of the parties. As the Supreme Court reaffirmed this year:

A class action, no less than traditional joinder (of which it is a species), merely enables a federal court to adjudicate claims of multiple parties at once, instead of in separate suits. And like traditional joinder, it leaves the parties’ legal rights and duties intact and the rules of decision unchanged.

Using the law actually applicable to a given claim is at the very least a "rule of decision," if not a full-fledged substantive right.

There are other possible flaws with the authors’ proposal (for example, there may well be Seventh Amendment implications). But the fact that Rosenberg and McCloud did not account for the clear dictates of class action law is reason enough that federal courts will likely steer far clear of it.