Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We study how machine learning techniques can be used to improve tax auditing efficiency using administrative data without the need of randomized audits. Using Italy’s population data on sole proprietorship tax returns and audits, our new approach addresses the challenge that predictions must be trained on human-selected data. There are substantial margins for raising revenue from audits by improving the selection of taxpayers to audit with machine learning. Replacing the 10% least promising audits with an equal number selected by our algorithm raises detected tax evasion by as much as 39%, and evasion that is actually paid back by 29%.