For those interested in the IRS audit coverage pre-COVID-19 here are graphs taken from the IRS 2019 Databook. The graphs depict just how low the audit coverage was before the IRS shut down for months. One can only imagine how low the coverage is at this point. Today’s post is one where we are adopting the philosophy that a picture is worth a thousand words. These pictures make this a very long post albeit one with few written words. For anyone who thinks it’s actually a good idea for the IRS to audit a reasonable number of returns to keep the system honest, the pictures here do not provide much encouragement.
read more...



compliance rates are likely falling, but we should also keep in mind that # of cases closed is only 1 of several imperfect metrics for judging the overall impact of IRS examination activities.
Dear Mr. Fogg,
This post is fine for what it is but it is incomplete. The table you selected ignores all of the compliance “touches” the IRS makes each year through its Automated Underreporter and Automated Substitute for Return programs. With those two, add another 1.9 million and 364,000, respectively. And that doesn’t include the 2 million math error notices. Compared to when I started practice, that is an incredible amount of contact with taxpayers. You might have better used Table 22 as a reference.
As more and more of tax reporting is automated one would expect the examination rate to go down. It also reflects well on the mostly competent and honest tax return preparers, the industry which has improved efficiency and accuracy of income reporting. Though the IRS still uses some technology that is out of date, it has a world class 1099 matching program. If the post is a lament for reduced opportunities for tax controversy practitioners, I’m afraid not many will be sympathetic. If it is a plea for more funding for the IRS, well, just like the rest of us, it has to do more with less. And anyway, government will always want more money.
Sincerely,
Joe Erwin
You are right that there is a lot more going on than just audit statistics. The programs you mention play an important role sometimes overlooked in the publication of audit statistics. Plus we all know what Mark Twain said about statistics. The lead slide dealing with large case exams, however, is not one where AUR, SFR and math error notices have any impact. The conclusion I draw on large case exams from the chart in the Data Book is one of significant collapse.