KPMG Exec Fined Too Much, Circuit Says

(CN) - A former consultant involved in the KPMG tax shelter fraud, called "the largest criminal tax case in American history" was fined $3 million too much, the 2nd Circuit ruled.

John Larson was one of three investment consultants convicted of creating fraudulent tax shelters for the New York accounting firm, having previously worked as the company's senior manager.

The government had initially charged 17 former KPMG executives with aggressively marketing and selling tax shelters, costing the U.S. Treasury $2 billion. In 2007, The New York Southern District Court threw out charges against all but four defendants, saying prosecutors violated their right to counsel.

District Judge Lewis Kaplan gave Larson a $6 million fine and sentenced him to 10 years in prison. Kaplan said the defendants were "motivated by greed" and their actions were "so raw, so brazen, so outrageous" that they clearly indicated criminal acts

Before a three-judge panel, the circuit found that the district court ignored case law in fining Larson.

"The jury found Larson guilty of 12 felony offenses, but made no findings as to the pecuniary gain or loss caused by his conduct," the circuit wrote. "Absent such gain or loss findings, the 'statutory maximum' fine Larson could receive was $3 million."

According to Supreme Court precedent, "any fact that increases the penalty for a crime beyond the prescribed statutory maximum must be submitted to a jury, and proved beyond a reasonable doubt."

The loss that Larson allegedly caused was not decided by a jury, and thus the court fined him beyond the maximum allowed by statute. The circuit vacated and remanded to district court Larson's fine.

It also upheld all the other convictions, including those against co-defendants Raymond Ruble and Robert Pfaff.

Subscribe to Closing Arguments

Sign up for new weekly newsletter Closing Arguments to get the latest about ongoing trials, major litigation and hot cases and rulings in courthouses around the U.S. and the world.

Email *

Submit