(Cristian Baitg/iStock)

After reaching a $101 million class action settlement to resolve lawsuits brought over a chemical spill that contaminated a West Virginia river, the plaintiffs lawyers asked a federal judge to grant them 30 percent of the fund as contingency fees.

The judge praised their work but found that fee request to be just too high.

“Even without accounting for fund size, the empirical literature clearly demonstrates that a 30 percent fee is higher than that awarded in the vast majority of class actions,” U.S. District Judge John Copenhaver of the Southern District of West Virginia wrote in a July order. “Courts have found through empirical analysis that larger common funds typically have smaller percentage fees.”

The empirical analysis Copenhaver referred to came from the findings of two leading academic reports — both cited in the opinion — that federal judges across the country have used for the past decade to guide them in decisions about attorney fees in some of the nation’s largest class action settlements.

New York University School of Law professor Geoffrey Miller and the late Theodore Eisenberg, a professor at Cornell Law School, wrote one of the studies, an updated version of which is set to be published this year. The second is a 2010 study conducted by Brian Fitzpatrick, a professor at Vanderbilt University Law School.

Both studies have provided critical assistance for federal judges, particularly when it comes to class action settlements of $100 million or more. The concern for those on the bench is how to award plaintiffs lawyers for their work without granting them excessive fees and leaving class members in the lurch.

“Judges do take the role seriously,” said William Rubenstein, a professor at Harvard Law School whose highly regarded “Newberg on Class Actions” has cited the Eisenberg/Miller and Fitzpatrick studies in his 11-volume treatise, alongside data he has used from a former publication called Class Action Attorney Fee Digest. “And they understand they’re a bulwark against excessive fees from the class members’ money.”

How to determine the exact amount has often been more art than science. In a webinar earlier this year hosted by the National Association of Legal Fee Analysis, U.S. District Judge David Herndon of the Southern District of Illinois, who has handled several of the nation’s largest mass torts and class actions, said a lot depends on the amount of recovery to the class.

“It just depends … on the case and what the benefit is that the lawyers have achieved by their work,” he said at the webinar, called “View from the Bench: Awarding Attorney Fees in Federal Litigation.” “If it’s reasonable, then you can approve the contingency, but if it’s pretty far out of whack maybe you’ve got to have the lawyers justify the difference or perhaps go with the lodestar. There are a lot of things to look at.”

And there are outside concerns as well. Judges have increasing scrutiny from appeals courts, which often take up the petitions of objectors to class action settlements, Rubenstein said.

“Public policy generally cautions against awarding too high a fee,” Copenhaver wrote in the West Virginia water case. “The court’s challenge is to award a fee that both compensates the attorneys with a risk premium on their skill and labor and avoids a windfall.”

Last month, plaintiffs lawyers in the case submitted a renewed motion for settlement approval that lowered their fee request to 25 percent — more in line with what Copenhaver had found was reasonable.

Brian Fitzpatrick.

Judges often look to previous case decisions, or their own experience, to determine what amount is appropriate to award lawyers in class actions. They also get a list of cases from the lawyers — but those often come with vested interests. For a long time, there was limited statistical data on what other judges had done. That’s where Fitzpatrick said he and the Eisenberg-Miller team tried to give judges a starting point.

“We tried to give the judges the full data instead of just leaving them at the whim of the cherry-picked cases the lawyers give them,” he said. “The judges don’t have to replicate what other courts have done, but they have the opportunity to stick within the mainstream of what their colleagues have done if they want it now that they have the power of empirical studies.”

Miller said he came up with the idea while serving as an expert witness in cases. When his first report with Eisenberg published in 2004, one year before the U.S. Class Action Fairness Act passed, the political atmosphere was rife with criticism about attorney fees in class actions. At the time, only one group had looked at the data — but it wasn’t really a controlled study.

“On the issue of fees, the data was there but hadn’t been developed,” he said. Eisenberg wasn’t an expert on class actions, Miller said, “but he was the leading person probably in the world who was doing empirical studies of legal material.”

Their report looked at published data of class action settlements from 1993 to 2002.

By the time of their second report in 2009, which expanded the data through 2008, Miller and Eisenberg had some competition. Fitzpatrick thought that their report, like those before it, relied too much on “ad hoc” data that focused primarily on bigger, published decisions.

“I really endeavored to find every last one to have the complete and representative picture,” he said.

He came up with a wider range of class action settlements within a shorter period of time — just 2006 and 2007.

Combined, both reports have been cited by judges more than 100 times, Miller said. And they often involve the biggest settlements in dollar amount, he said.

“The issue is that there aren’t as many cases,” he said. “There’s less data. And that puts an additional premium on getting what data there is, so that’s one reason judges look to this research in big cases.”

Another came in 2012, he said, when U.S. District Judge Lee Rosenthal of the Southern District of Texas, the former chairwoman of the Judicial Conference Committee on Rules of Practice and Procedure, endorsed both studies in a case called In re Heartland Payment Systems Customer Data Security Breach Litigation: “District courts increasingly consider empirical studies analyzing class-action-settlement fee awards to set the appropriate percentage benchmark or to test the reasonableness of a given benchmark,” she wrote. “Using these studies alleviates the concern that the number selected is arbitrary.”

Economies of Scale

Both studies have come out with slight differences in their specific findings. But they came to the same conclusions: The vast majority of judges award fees based on a percentage of the total settlement amount — then cross-check that amount against the total number of hours the lawyers billed multiplied by the hourly rate, referred to as the lodestar. There’s a good reason for that trend.

“It’s economies of scale,” Miller said. “Judges understand that to get a $1 billion settlement is not 1,000 times harder for an attorney to get a $1 million settlement. It’s a lot harder, but not 1,000 times harder.”

Herndon, in the webinar, said that’s just common sense.

“If they got a third of $1 billion, and compared to their lodestar, it would be an astronomical per hour figure,” he said. “There’s some common sense in doing something like that, and I don’t really have a particular feeling one way or the other, but I think there’s certainly authority in the law for doing it.”

In fact, many judges who cite the Eisenberg-Miller and Fitzpatrick reports look specifically to the data as it pertains to the size of the settlement in front of them and what the case is about.

But Fitzpatrick questioned whether judges were doing the right thing in lowering the percentages as the settlements get bigger.

“I think the judges are responding to perception when they do that and they’re not responding to good economic policy analysis,” he said. “Because why would we want to punish lawyers with lower percentages for getting their clients more money?”

Not all judges agree with the conclusions made by the professors, who sometimes go up against each other as paid experts in individual cases. In a $415 million settlement of “no poach” claims involving high-tech workers, U.S. District Judge Lucy Koh of the Northern District of California weighed Fitzpatrick’s report against the Eisenberg/Miller study in awarding $40 million in fees. In that case, Fitzpatrick was a paid expert for the lead plaintiffs attorneys, while Rubenstein cited the Eisenberg-Miller report in a declaration filed on behalf of one of the lead firms that had submitted a separate fee request.

“The court finds the Eisenberg & Miller study more persuasive than the Fitzpatrick study,” Koh wrote in a 2015 order, concluding that the “length and large sample size of the Eisenberg & Miller study suggest that its results are entitled to greater weight.”

Fitzpatrick said he’s working on an updated report, likely to be drafted next year. “Whenever I hear from these judges, they say the same thing: We love your study but we need more recent data,” Fitzpatrick said. “So that’s what I’m trying to give them.”

But gathering the data takes a lot of time and money, he said. He’s hired research assistants to code all the data.

The latest Eisenberg-Miller report, co-authored with research scholar Roy Germano at NYU’s law school, uses data through 2013. Without Eisenberg, who died in 2014, Miller said he’s not certain he’ll keep publishing the report.

“I don’t think I’ll do it anymore,” he said. “It is a lot of work.”