IRS Audits Blacks at substantially higher rates than others

[laughing dog]

A study published Monday shows that IRS algorithms target blacks for auditing at much higher frequencies even after controlling for other variables. Remember that race is not part of the IRS tax forms.

The paper, published by Stanford’s Institute for Economic Policy Research, said that despite the IRS’s “race-blind audit selection,” Black taxpayers are audited 2.9 to 4.7 times more often than non-Black taxpayers. However, the research did not suggest that the disparity is a result of one group of people evading taxes more than another but rather that it may be a result of the computer algorithms the IRS uses for selection.

“Using counterfactual audit selection models, we find that maximizing the detection of underreported taxes would not lead to Black taxpayers being audited at higher rates,” the paper reads.

The study found that the largest disparity between the groups was among those claiming the earned income tax credit (EITC), which helps low- to moderate-income workers and families get a tax break, according to the IRS.

“Even when such algorithms are formally neutral with respect to protected characteristics like race, there is widespread concern that they can disproportionately burden vulnerable groups,” the economists wrote.

(source : Blacks much more likely to be audited by the IRS )

Note also that these audits are more likely to be middle and low income earners (especially EITC eligible filers).

I don't think this result is something that should be shrugged off as "whaddya going to do" because algorithms are based on decisions made by people.

 

[Jimmy Higgins]

It seems one of the bigger issues is the algorithms are trying to catch people over-refunding verses under-reporting income. Both would seem important, though over-refunding via too many dependents or what not might be easier to prove than proving someone under-reported income.

 

[Oleg]

A study published Monday shows that IRS algorithms target blacks

Really? Of all the times I've cursed and did my taxes, I cannot recall ever having to put my race or sex on the 1040.

 

[Jimmy Higgins]

A study published Monday shows that IRS algorithms target blacks

Really? Of all the times I've cursed and did my taxes, I cannot recall ever having to put my race or sex on the 1040.

And yet blacks are audited about 3+ times more often. Blacks aren't "targeted", but they are getting "audited".

 

[Oleg]

A study published Monday shows that IRS algorithms target blacks

Really? Of all the times I've cursed and did my taxes, I cannot recall ever having to put my race or sex on the 1040.

And yet blacks are audited about 3+ times more often. Blacks aren't "targeted", but they are getting "audited".

Is it because they're black, or does socio-economics provide a more likely and less inflammatory explanation?

 

[ZiprHead]

A study published Monday shows that IRS algorithms target blacks

Really? Of all the times I've cursed and did my taxes, I cannot recall ever having to put my race or sex on the 1040.

And yet blacks are audited about 3+ times more often. Blacks aren't "targeted", but they are getting "audited".

This is an example of systemic racism, a system that doesn't purposely target black people but the effect is the same.

 

[Jimmy Higgins]

A study published Monday shows that IRS algorithms target blacks

Really? Of all the times I've cursed and did my taxes, I cannot recall ever having to put my race or sex on the 1040.

And yet blacks are audited about 3+ times more often. Blacks aren't "targeted", but they are getting "audited".

Is it because they're black, or does socio-economics provid a more likely and less inflammatory explanation?

Long story short, it is based on what the IRS is targeting:

In contrast, in these models, certain policies tend to increase the
audit rate of Black taxpayers: (1) designing audit selection algorithms to minimize
the “no-change rate”; (2) targeting erroneously claimed refundable credits rather
than total under-reporting; and (3) limiting the share of more complex EITC returns
that can be selected for audit.

Now adults and professionals who look at the results, would usually step back and say... gee... 3x more audits? Is the IRS getting paid or are the audits useless? And from there, determine whether expanding the audits in these areas makes sense or whether they need to modify the algorithms to reduce false positives.

 

[Oleg]

A study published Monday shows that IRS algorithms target blacks

Really? Of all the times I've cursed and did my taxes, I cannot recall ever having to put my race or sex on the 1040.

And yet blacks are audited about 3+ times more often. Blacks aren't "targeted", but they are getting "audited".

This is an example of systemic racism, a system that doesn't purposely target black people but the effect is the same.

If it doesn't purposely target black people, it's not systemic racism. Unless systemic racism is just a meaningless trope.

 

[TomC]

A study published Monday shows that IRS algorithms target blacks

Really? Of all the times I've cursed and did my taxes, I cannot recall ever having to put my race or sex on the 1040.

And yet blacks are audited about 3+ times more often. Blacks aren't "targeted", but they are getting "audited".

This is an example of systemic racism, a system that doesn't purposely target black people but the effect is the same.

Is it?
Or is it an example of wokesters attributing racism without explanations?
Because it matches their preconceived notions.
"Black people are victims of institutional racism because government is bad. Any time anything happens that a black person doesn't like it's obviously a result of white racists."
Tom

 

[ZiprHead]

Here's one definition of systemic racism:

Racism is not always conscious, explicit, or readily visible—often it is systemic and structural. Systemic and structural racism are forms of racism that are pervasively and deeply embedded in systems, laws, written or unwritten policies, and entrenched practices and beliefs that produce, condone, and perpetuate widespread unfair treatment and oppression of people of color,

https://www.healthaffairs.org/doi/10.1377/hlthaff.2021.01394

The rest of the article is about systemic racism in healthcare.

 

[Jimmy Higgins]

FFS!

Blacks are apparently audited much more often than whites. The report indicates the reasons why. None of them include bigoted hate. They do indicate it is systemic in how the system is designed to look for potential returns that need audits. It appears the system needs to be revisited so the outcomes manage to capture more people who need auditing... (the actual target) not one race or the other.

 

[Don2 (Don1 Revised)]

A study published Monday shows that IRS algorithms target blacks

Really? Of all the times I've cursed and did my taxes, I cannot recall ever having to put my race or sex on the 1040.

And yet blacks are audited about 3+ times more often. Blacks aren't "targeted", but they are getting "audited".

Is it because they're black, or does socio-economics provide a more likely and less inflammatory explanation?

It's odd how conservolibertarians like you are always complaining about taxes, tax enforcement, and tax audits, but suddenly it's shown objectively that blacks ought to be the ones complaining more often and it's all "that's inflammatory!!111!"

 

[Elixir]

A study published Monday shows that IRS algorithms target blacks

Really? Of all the times I've cursed and did my taxes, I cannot recall ever having to put my race or sex on the 1040.

And yet blacks are audited about 3+ times more often. Blacks aren't "targeted", but they are getting "audited".

Is it because they're black, or does socio-economics provide a more likely and less inflammatory explanation?

It's odd how conservolibertarians like you are always complaining about taxes, tax enforcement, and tax audits, but suddenly it's shown objectively that blacks ought to be the ones complaining more often and it's all "that's inflammatory!!111!"

Not to detract from Don’s point, but I think Ollie is right that IRS discriminates against poor people, not skin color. But as it happens, it turns out to be almost the same thing. A deeper look into the data would likely reveal that blacks are audited 3x the rate of whites because they are 3x more likely to fall into the income range that is easiest/ most propitious to go after. Repugs and libbertards seem dead-set against enabling the IRS to go after rich people.

 

[TV and credit cards]

Note, federal tax returns taking standard deduction is straight forward and easy to see mistakes triggering potential audit. It's when a taxpayer itemizes and utilizes various Schedules to complete their taxes, interpreting the rules starts to gray.
Also note, the study utilizes Bayesian Improved First Name Surname Geocoding (BIFSG) to claim their audit rates, this is in which no voluntary racial data is collected.

People collecting EITC are getting busted for claiming it when they are not entitled. People who claim EITC are poor to moderate income.

The study found that the largest disparity between the groups was among those claiming the earned income tax credit (EITC), which helps low- to moderate-income workers and families get a tax break, according to the IRS.

“Even when such algorithms are formally neutral with respect to protected characteristics like race, there is widespread concern that they can disproportionately burden vulnerable groups,” the economists wrote.

There's no gray area with regards to EITC. You're either entitled to it based on income or you're not.

 

[Jimmy Higgins]

Note, federal tax returns taking standard deduction is straight forward and easy to see mistakes triggering potential audit. It's when a taxpayer itemizes and utilizes various Schedules to complete their taxes, interpreting the rules starts to gray.
Also note, the study utilizes Bayesian Improved First Name Surname Geocoding (BIFSG) to claim their audit rates, this is in which no voluntary racial data is collected.

People collecting EITC are getting busted for claiming it when they are not entitled. People who claim EITC are poor to moderate income.

The study found that the largest disparity between the groups was among those claiming the earned income tax credit (EITC), which helps low- to moderate-income workers and families get a tax break, according to the IRS.

“Even when such algorithms are formally neutral with respect to protected characteristics like race, there is widespread concern that they can disproportionately burden vulnerable groups,” the economists wrote.

There's no gray area with regards to EITC. You're either entitled to it based on income or you're not.

As far as I can tell, there is no indication that those who were audited were found doing anything wrong. Which is the problem.

 

[laughing dog]

A study published Monday shows that IRS algorithms target blacks

Really? Of all the times I've cursed and did my taxes, I cannot recall ever having to put my race or sex on the 1040.

Do you have a point other than show you didn't bother to read the OP or the link?

 

[laughing dog]

A study published Monday shows that IRS algorithms target blacks

Really? Of all the times I've cursed and did my taxes, I cannot recall ever having to put my race or sex on the 1040.

And yet blacks are audited about 3+ times more often. Blacks aren't "targeted", but they are getting "audited".

This is an example of systemic racism, a system that doesn't purposely target black people but the effect is the same.

If it doesn't purposely target black people, it's not systemic racism. Unless systemic racism is just a meaningless trope.

Whether or not it is "systemic racism", it is exactly the kind of "unintended" result that seems to pop more often than it should. Moreover, since federal income tax collection is based on voluntary compliance, such outcomes create distrust which leads to less voluntary compliance.

 

[Loren Pechtel]

A study published Monday shows that IRS algorithms target blacks for auditing at much higher frequencies even after controlling for other variables. Remember that race is not part of the IRS tax forms.

The paper, published by Stanford’s Institute for Economic Policy Research, said that despite the IRS’s “race-blind audit selection,” Black taxpayers are audited 2.9 to 4.7 times more often than non-Black taxpayers. However, the research did not suggest that the disparity is a result of one group of people evading taxes more than another but rather that it may be a result of the computer algorithms the IRS uses for selection.

“Using counterfactual audit selection models, we find that maximizing the detection of underreported taxes would not lead to Black taxpayers being audited at higher rates,” the paper reads.

The study found that the largest disparity between the groups was among those claiming the earned income tax credit (EITC), which helps low- to moderate-income workers and families get a tax break, according to the IRS.

“Even when such algorithms are formally neutral with respect to protected characteristics like race, there is widespread concern that they can disproportionately burden vulnerable groups,” the economists wrote.

(source : Blacks much more likely to be audited by the IRS )

Note also that these audits are more likely to be middle and low income earners (especially EITC eligible filers).

I don't think this result is something that should be shrugged off as "whaddya going to do" because algorithms are based on decisions made by people.

So once again they can't actually find discrimination and resort to using disparate impact as "evidence".

And note that the objective of the IRS isn't to maximize detection of underreported taxes, but to maximize the detection:cost ratio. In practice this means they go after the easiest ones to detect, not the biggest ones. The poor are likely to have done their taxes by hand and thus have math errors the computer can detect and they're less likely to realize the stupidity of any sort of tax cheating the computer can catch.

Meanwhile the more careful crowd goes virtually ignored because the IRS simply doesn't have the resources to chase them down.

 

[Loren Pechtel]

Now adults and professionals who look at the results, would usually step back and say... gee... 3x more audits? Is the IRS getting paid or are the audits useless? And from there, determine whether expanding the audits in these areas makes sense or whether they need to modify the algorithms to reduce false positives.

There's something that needs to be understood here: "Audit" covers a great range of actions. At one end we have the TCMP audits from hell, at the other end we have what happened to me one year--got a letter from them saying I had put down a wrong date, the correct date is xxxx, we have adjusted the tax calculations, you owe us [trivial amount] in interest. Since an amount this small isn't worth collecting we are writing it off, do not submit a payment. I'm not sure exactly where the error was because it wasn't worth hunting down over $0.

Most IRS audits are a letter about something the computer didn't like and are resolved either by paying the new numbers the IRS gives or amending your return to fix the actual error.

 

[Loren Pechtel]

If it doesn't purposely target black people, it's not systemic racism. Unless systemic racism is just a meaningless trope.

It's possible for something to not "purposely" target a group and yet be discriminatory. We saw a lot of that with the civil rights movement--companies doing things like putting height requirements on jobs where height doesn't matter--but they picked a height that would exclude few men but most women. Be very leery of changes proposed after a company has been smacked for discrimination.