New frontiers in "equity" -- the push for a color-conscious IRS

The “equity” crowd isn’t aiming for fairness and impartiality — i.e. equity. It’s goal is to impose equal racial outcomes in every area of American life, which requires a huge amount of partiality in favor of blacks.

Taxation is the latest battleground for the equity crowd. It wants the IRS to collect race data in the name of equity.

The first thing to notice here is the tension between this demand and the claim that America is “systemically racist.” If the equity crowd really believed this claim, the last thing it would want is for the government’s tax collecting arm (which is often controlled by Republicans) to know which taxpayers are black.

Actually, this is the last thing any of us should want. The only color the IRS should worry about is green.

But the left understands that the U.S. government is systemically partial to blacks. Democratic administrations favor their interests pretty much across the board. Republican administrations rarely dare reverse policies that favor blacks.

This is why we’re hearing calls for the IRS to collect racial data. The left wants to enlist the IRS in its quest for “equity.” If the government knows the race of taxpayers, it can easily favor preferred minority groups, just as it does across a wide range of programs including student loan debt relief.

Advocates of color conscious tax reporting seem to admit that this is their purpose:

Some scholars argue that the IRS’ lack of appetite for collecting racial and demographic information hinders its ability to make informed decisions that would better serve marginalized groups, especially those with low incomes and/or in certain areas of the country.

Marginalized groups with low income are already well served by federal tax policy. They pay much less tax than whites, thanks to the progressive nature of the tax code.

However, proponents claim that identifying taxpayers by race will shine light on discrimination in tax enforcement. Says one leading proponent:

The IRS should be tracking tax enforcement actions by race in order to identify potential racial disparities and also correct them.

Are there such disparities? A study by Stanford University researchers found that blacks have their tax returns audited to a disproportionate degree. Based on a review of 148 million tax returns and nearly 800,000 audits, the study concluded that black taxpayers are audited at least 2.9 times the rate of non-black taxpayers.

Because tax returns contain no information about the individual’s race, the researchers had to use proxies for race, such as first names, last names, and U.S. Census data to predict the likelihood that given taxpayers are black. To cross-check these predictions, the team used voter registration records from North Carolina, where citizens were required to self-report race and ethnicity when they registered.

Clearly, this methodology provides only an approximation of the racial characteristics of those whose taxes are audited. However, based on my experience as a lawyer with such methodologies and given the size of the disparity found, I think it’s fair to conclude that blacks are, in fact, audited to a disproportionate extent.

Why is that? The researchers acknowledged that it’s probably not the result of racial targeting. The IRS doesn’t know the race of tax filers. It selects the people it audits using algorithms that don’t take race into account. The disparities result from using these race-neutral algorithms, not from targeting blacks.

Critics claim the algorithms used by the IRS are “dated.” They may be old, but that doesn’t mean there’s anything wrong with them.

The problem isn’t datedness. If old algorithms were resulting in disproportionate auditing of white taxpayers, the equity crowd would have no complaint.

But it turns out that the old algorithms disproportionately identify blacks for audits:

Looking at the data, the team determined the IRS has focused largely on a specific set of mistakes taxpayers make, including mistakes that affect their eligibility for refundable credits.

These mistakes are common among Black taxpayers, despite Black taxpayers being less likely to make the highest dollar value mistakes overall.

I see no reason to change algorithms that are facially neutral as to race merely because they disproportionately identify blacks for audits. To do so smacks of partiality. Furthermore, to the extent that blacks make a disproportionately large number of one type of mistakes, it’s far from clear that they well fare better under a system that focuses on different kinds of mistakes. (What we really need is radical reform of our tax system to eliminate the complexities that are causing taxpayers of all races to make so many mistakes.)

In any case, if there’s something wrong with the algorithms, the solution is to change them — not to have taxpayers identify their race to the IRS.

This is particularly true, I think, given the Biden administration’s plan to hire a large number of new agents — possibly as many as 87,000 — for the purpose of closing an estimated $600 billion annual gap between taxes owed and taxes unpaid. This entails a big increase in the number of audits.

As the IRS audits more and more Americans, it becomes more and more imperative that Americans have confidence that no racial group is being targeted. If the IRS is able to identify taxpayers by race, Americans are unlikely to have that confidence — nor is it clear that they should.