A D.C. Circuit panel has raised concerns with a survey of lawyers' rates that a lower court used to pare millions from an attorney fee award in a decision that legal aid and public interest groups fear may undercut their fee awards and ultimately reduce the number of clients they can take on.

The dispute hinges on a new attorney fee matrix put forward by the government that is based on a survey of hourly rates billed by lawyers throughout Washington, D.C., Maryland, Virginia and West Virginia who practice in various legal arenas, including in litigation, wills, bankruptcy cases and real estate closings, among other practices.

A D.C. district court relied on that matrix two years ago to shave just under $3 million from the nearly $10 million in fees and costs a class requested to pay their attorneys after winning a long-running case against the nation's capital over violations of federal and local disability laws. D.C. had championed the new data, as it argued the government's updated rates provide a more reliable and accurate measure of the current market, while the iteration the class prefers is three decades old.

In the class' appeal, it told the D.C. Circuit that complex litigation attorneys in D.C. shouldn't be paid from a dataset that includes non-litigation lawyers hailing from more rural areas that charge far less hourly.

And more than a dozen nonprofits that rely on fee-shifting to support their caseloads — including the National Women's Law Center, AARP and the Animal Legal Defense Fund — lodged their support for the class' challenge, as they said in an amicus brief that awarding below-market rates curbs the number of clients a public interest organization can afford to represent.

At oral arguments last week, the panel also voiced concerns with the survey's widely cast net, as Judge Merrick Garland said he couldn't understand why a lower court would use a broad sweep of rates to pay a set of complex litigation lawyers based out of D.C. whose hourly rates would differ significantly from others included in the survey.

"If you have a dataset that intentionally includes rates that are neither complex nor District of Columbia, I don't understand how we can give any weight at all to that dataset," Judge Garland remarked.

And Judge David Tatel echoed his colleague's thoughts.

"It can't be that by including huge numbers of lawyers that are not in district and legal work that's not only not complex but not litigation, that that could conceivably average out to something that's useful to us," he said.

The suit originated in 2005 when parents of then-preschool-aged children challenged D.C.'s failure to provide special education services for their kids. After they won, they asked for $9.8 million to pay their counsel for more than a decade of work, but the lower court ultimately trimmed the fees and costs to $6.9 million to align with averages from the survey.

Assistant Attorney General Lucy Pittman argued for the district in the hearing, and she conceded that the data spans a range of legal specialties, including wills and estates, where attorneys' rates would clock in on the lower end of the pay scale. And the judges highlighted the fact that the data includes lawyers' hourly rates in rural parts of Maryland, Virginia and West Virginia, which would also pull the average down.

But Pittman argued that the bankruptcy attorneys' fees were also factored in, and as they lie on the other end of the spectrum, she said they balance out the average.

However, Judge Garland said two wrongs don't make a right.

"You don't get a statistically correct [result] by taking wrong numbers on one side and wrong numbers on the other, and hope that just randomly, or by accident, they average out to something useful," he said.

The U.S. also waded into the battle to support the district, as U.S. Department of Justice attorney Charles W. Scarborough argued at the hearing that the data isn't perfect, but emphasized it's a "reasonable effort" and a "statistically reliable way" to update the decades-old matrix.

Judge Tatel jumped in to say he understands the government's push to set out a fresh data set — which he said is clearly "brand new" and "much more up to date" — however, he's still not certain it's useful here.

"The question is, is the data relevant?" he asked.

--Editing by Pamela Wilkinson.

Have a story idea for Access to Justice? Reach us at accesstojustice@law360.com.