I'll admit - the 415:1 was pulled from an article detailing information from 2024 but the main point isn't the actual value but the fact that it's more than 1:1. When the IRS receives more funding the US government gets more money than what it is budgeting - this doesn't scale to infinity, at some point you'll have nearly complete auditing capture and more budget will just be burning money but we're no where near that point.
Putting money into the IRS is basically a free money printer for the US government and it's only deep corruption that keeps it so poorly funded.
They already have another money printer that they’re perfectly happy to rely on, at least for the time being.
There are substantial indirect costs not accounted for in that ratio. Anywhere close to 1:1 is a large net loss to the government. Your mental model of the cost effectiveness of audits completely ignores large second-order effects.
The Federal government has a century of empirical data on this. They set their targets accordingly, which as a heuristic is roughly optimal at around 10:1. This may not be intuitive to you. It wasn’t to me either until I worked at a Federal audit agency. Most of it actually makes sense once you understand the bigger picture.