Member-only story
Quick Success Data Science
Exposing Bias in IRS Audits Using Python and Plotly Express
Why they target the poor and how to fix it
According to IRS chief Charles Rettig, the US Treasury loses $1 trillion in unpaid taxes every year. Most of these unpaid taxes are the result of evasion by the wealthy and large corporations. To address this, the Inflation Reduction Act of 2022 included almost $80 billion for Internal Revenue Service modernization and increased tax enforcement.
Despite the stated goals, news agencies began reporting that as many as 87,000 new auditors and agents would be hired and that they would go after the poor and middle class, rather than rich individuals and corporations. As the debate raged on cable news, opposing sides quickly grouped into partisan camps, with each accusing the other of spouting misinformation and “fake news.”
If you’re a data scientist, you’re not at the mercy of the pundits. You have the power to answer these types of questions for yourself, using facts, Python, and a few third-party libraries. In this Quick Success Data Science project, we’ll use IRS data to map some of the agency’s historical audit behaviors. You may be surprised at the results.
