One challenge defendants and their counsel face early in defending a putative class action filed in state court, assuming they would prefer to litigate in federal court, is how to show that the $5 million amount in controversy requirement is satisfied.  This must be done quickly so that the removal can be timely filed within 30 days, and it often takes considerable time to find the right people to do the analysis, figure out what the relevant data is, analyze various iterations of data and then decide how best to present this issue to the court.  The courts of appeals have agreed that the burden of proof is on the defendant to show that the amount in controversy is satisfied, although there is some variation in how the applicable standard is articulated.

A recent Northern District of Ohio decision demonstrates the importance of tailoring the defendants’ data to the definition of the proposed class, where possible, as opposed to using more generic data that may be more easily retrievable but is not specific to the proposed class.  Andrews v. Nationwide Mutual Insurance Co., 2011 U.S. Dist. LEXIS 124737 (N.D. Ohio Oct. 26, 2011) is one of a series of class actions recently filed seeking to require life insurance companies to search for information about whether their insureds have died (typically by using the Social Security Administration’s Death Master File) and then make more proactive efforts to pay proceeds where a claim has not been submitted.  (For more about these kinds of cases, see the prior posts I did on June 2, 2011 and July 13, 2011.)  The complaint sought injunctive and declaratory relief, and also asserted claims for breach of the implied covenant of good faith and fair dealing and unjust enrichment.  The only issue in dispute on jurisdiction was the amount in controversy. 

Nationwide’s notice of removal alleged “three components of damages: (1) $826,000, which constitutes the face value of active life insurance policies for which the insured has been determined to be deceased, (2) $1,228,000, which constitutes the value of lapsed insurance policies during the past 15 years for which the insured has been determined to be deceased; and (3) the $10,200 annual cost to conduct yearly searches of the Death Master File (‘DMF’) for active life insurance policyholders and monthly searches for all lapsed policies.”  Id. at *3-4.  Nationwide argued that “the injunction plaintiffs seek is indefinite and perpetual and, as such, the amount in controversy is satisfied by category three alone.”  Id. at *4.

The court found this showing insufficient in large part because the data was not adequately tailored to the putative class as defined in the complaint:

Upon review, the Court finds that defendants have not established a $5 million amount in controversy by a preponderance of the evidence. Defendants provide the affidavit of Jeffrey Stein, who avers that defendants ran searches of their active and lapsed policies against the DMF. The results of these searches indicated that there are approximately 230 active policies worth $826,000 for which defendants received an “exact or near exact match” and for which defendants could not locate the policy beneficiary. In addition, there are approximately 17 lapsed policies worth $1,228,000 that fit the same criteria. Defendants ask the Court to include these dollar figures in the amount in controversy. These dollar figures, however, are derived from searches of defendants’ entire book of business. Thus, these figures do not accurately represent the interests at stake in this case, as the number of policies at issue is far less than defendants’ entire book of business. For example, there is no indication as to how many of the 230 active policies and 17 lapsed polices were held by class members. Nor is there any indication as to the value of the class members’ policies.

In addition, the Court rejects defendants’ argument that the cost of running searches ad infinitum would itself exceed the jurisdictional amount. Defendants provide evidence that the cost to run the searches is $10,200 per year. Defendants’ argument ignores the simple fact that the injunction could not, by definition, run ad infinitum. Count one, which seeks mandatory injunctive relief, asks the Court to order defendants to make reasonable inquiries as to the “life-status of the Class Members.” Similarly, count two asks that the Court declare that defendants must pay death benefits to “Class Members … without first requiring further notice of death.” On the face of the complaint, the injunctive and declaratory relief is requested only on behalf of “Class Members.” The class is defined generally as individuals who have policies that are “currently in force” or have been “wrongfully canceled” and who held such policies “within the period of time that commenced 15 years prior to the filing” of this lawsuit Construing the class definition as broadly as possible, the injunction could only last as long as the youngest person in the class is alive. Once the youngest person is deceased, the injunction would necessarily expire as there would be no more records to search or benefits to pay.

Id. at *7-10 (footnotes omitted).

It seems likely here that at least some of Nationwide’s data could have been tailored more closely to the proposed class.  But it may very well be that if the data had been tailored specifically to the proposed class, Nationwide would have had no hope of reaching the $5 million threshold, and thus they may have just been giving this the old college try.  What I think is critical here is for insurers and their counsel to understand that thoroughly developing the amount in controversy data to be used in a CAFA removal requires substantial time and effort.  In some cases it is simply not possible to tailor the data to the proposed class as defined in the complaint because the insurance company simply does not maintain the right kind of data, but you can probably get pretty close.  That may take several meetings with the right people knowledgeable about the company’s data, and working through several different iterations of data that you might want to use, in order to get the best possible data and accompanying argument to present to the court.  From the outside (and in-house) counsel perspective, gaining a deep understanding of how the client maintains the data can be essential.  Don’t underestimate the amount of time and effort that is required for this, and don’t assume that a judge will accept some kind of rough, back of the envelope calculation.  That might fly where it is clear that the case involves much more than $5 million, but it might not where, as in this case, it was a closer call.