MIT Scientists Create Algorithm For Blowing Up Tax-Evading Corporations’ Spot

Photo by Margaret Burdge

Swiss bank accounts are a trope nowadays, in equal parts for their world-renowned secrecy and their use by the likes of Bond villains and hedge-fund managers (two groups that aren’t exactly mutually exclusive). But tax shelters exist all across the globe in various forms, from offshore subsidiaries to financing agreements. Notoriously hard to sniff out, shelters used by individuals and corporations cost federal and state governments approximately $184 billion each year, according to the U.S. Public Interest Research Group.

Enter MIT.

A recent paper—titled, “Tax Non-Compliance Detection Using Co-Evolution of Tax Evasion Risk and Audit Likelihood” and authored by three MIT computer scientists and two from the Mitre Corporation, a technology research and development nonprofit—has demonstrated how artificial intelligence can be used to distill thousands of pages of tax law into one beautiful algorithm aimed at rooting out scofflaws.

“When the IRS pursues a tax evasion scheme and changes the tax law or audit procedures, the tax evasion schemes evolve and change into undetectable forms. The arms race between tax evasion schemes and tax authorities presents a serious compliance challenge,” the paper says. Currently, the IRS combs through pre-existing data from filed tax returns to search for tax shelter schemes. Instead of data-mining, these computer scientist have proposed rule-mining.

From the New York Times’ explanation of their methodology:

First, the researchers translated tax regulations governing partnerships, a growing source of tax trickery, into source code. Then they rendered the transactions underpinning a questionable shelter known as “installment-sale bogus optional basis,” or Ibob, as a series of codes. The Ibob shelter artificially inflates the basis value of an asset on a tax return to wipe out taxable gains when that asset is sold. While some of Ibob’s individual transactions are perfectly legal, the collective result is a bogus deduction.

Next, the researchers mapped out in code the tangle of entities that make up typical partnerships. The results flagged specific combinations of transactions and partnership structures that were likely to produce the Ibob dodge.

Taking notes? There’ll be a quiz.

The scientists admit in the paper’s conclusion that there’s still some fine-tuning to be done. Still, the findings are a promising development in the ongoing struggle against tax evasion of the most sophisticated sort.