IRS Faulted for Efforts to Curb Phony Tax Refunds to Prisoners

After six years of congressionally mandated effort, the Internal Revenue Service still lacks solid procedures for identifying tax refund fraud committed using prisoner Social Security numbers, a watchdog found.

The number of such fraudulent returns flagged by the agency rose from 37,000 valued at $166 million in calendar year 2007 to 137,000 valued at $1 billion in 2012, according to the third in a series of audits by the Treasury Inspector General for Tax Administration, released on Tuesday. That is a sign of a growing problem, the watchdog said.

"Since TIGTA first began documenting this fraudulent activity four years ago, refund fraud committed by prisoners has grown to become a billion-dollar problem," said J. Russell George, Treasury Inspector General for Tax Administration. Calling for more to be done, George said, “it is incumbent upon the IRS to act aggressively to prevent tax fraud wherever it occurs, particularly behind bars."

Specifically, TIGTA noted that the IRS has not yet shared fraudulent prisoner tax return information with federal or state prison officials. The agency’s mandatory annual prisoner fraud reports to Congress “are not timely and … the reports do not address all aspects of prisoner fraud,” auditors said. And the IRS has delayed fulfilling a past recommendation that it ensure that all tax returns filed using a prisoner Social Security number are assigned a prisoner indicator.

TIGTA urged the commissioner of the agency’s Wage and Investment Division to address six recommendations, four of which were accepted.

The IRS did not agree to develop a process to identify other tax returns that have the same characteristics as confirmed fraudulent prisoner returns. It said the methodology used in its annual report to Congress on all known fraudulent returns is consistent over the years, and that the alternative methods recommended by TIGTA are not reliable enough to conclude that the returns were filed by inmates.

The tax agency also rejected a suggestion that it correct a computer programming error that caused omission of a prisoner indicator on 3,139 returns, saying it affected only 3 percent of transcribed paper tax returns, which are further cross-checked.

(Image via Oleksiy Mark / Shutterstock.com)