Sophia Bagley
One criticism we’ve seen of Elon Musk’s Department of Government Efficiency (DOGE) is that it could actually worsen the federal government’s budget deficit. Because DOGE is involved in laying off IRS workers, even some free-market conservatives complain that it will result in less legitimate tax revenue being collected.
Certainly, the marginal IRS worker will bring in some additional revenue—likely more than what the government pays for their salary. But is that the right metric for hiring more IRS agents? Of course not. This is an area where the government’s fiscal position and a broader conception of efficiency could be in tension.
IRS workers seek to eradicate false negatives—people not paying taxes even though they have taxes due. But the way IRS agents obtain that revenue is usually through conducting audits of taxpayers, many of whom will be investigated despite paying the correct taxes (false positives). That means a ton of time and energy is wasted on fruitless audits, including administrative stress for the individuals investigated.
As my Cato colleague Chris Edwards has pointed out, ramping up audits without reform risks inflicting serious inefficiencies. Ideally, we should reduce taxpayer errors and gaming by simplifying the tax code and modernizing IRS systems. Despite claims that the rich and corporations are cheating at vast levels, the IRS’s own “tax gap” estimate has remained stable for decades relative to gross domestic product.
So, yes, the IRS needs to audit to ensure compliance. But examining the effects of more agents on revenue alone paints a partial picture. More aggressive enforcement imposes growing costs on compliant businesses and individuals through legal fees, uncertainty, and in some cases, business closures. One study found that audited firms are more likely to shut down after being audited.
The Congressional Budget Office has shown that enforcement has diminishing returns. A 2020 report notes that adding $20 billion to IRS enforcement could bring in $61 billion, but a second $20 billion would yield only $42 billion more. A more recent report concurs that the return on investment from IRS enforcement activities “drops by 10 percent for every 10 percent increase in spending for enforcement and related activities over a base amount of about $10 billion.” These returns should be weighed against the private-sector costs and broader economic disruptions from heavy-handed enforcement.
Weighing these trade-offs is ultimately a technocratic question, over which we libertarians have little expertise. The point is: Looking at DOGE’s IRS employee cuts through the lens of lost revenue alone will exaggerate the downsides. Shrinking the workforce compels the agency to prioritize audits more carefully. Historically, audit precision improves when resources are limited, with no-change rates falling as a result. And many people who don’t need to be audited will avoid them.