Contrary to popular belief, the IRS has massive funding. They just got injected with $80B to hire 87k agents over 10 years. I cannot imagine any private entity operating this inefficiently.

This claim is not correct in several ways. The $80B was not earmarked for hiring agents; that was an additional $80B for the entire agency spread over 10 years. While some of that money might be used to hire additional agents, it will also be used to upgrade old technology (the IRS still uses software from the 1950s), make fixed capital investments that have been long-deferred, regulate third-party tax preparers, provide better customer service, crack down on tax evasion and improve enforcement.

Additionally, this funding needs to be set in the context of an agency whose budget has shrank 20% over the last decade and whose staffing is set to further shrink through attrition and retirements. The IRS Commissioner testified that the IRS would have to hire 52,000 employees over the next 10 years just to keep staffing at current levels [1].

Finally, the Treasury has argued that this $80B investment would (conservatively) lead to an additional $700B in tax revenue over the next decade, so this spending ends up being entirely revenue-positive [2].

[1] https://www.irs.gov/newsroom/written-testimony-of-charles-p-...

[2] https://home.treasury.gov/system/files/136/The-American-Fami...

Edit: since this is Hacker News, I thought it worth pointing out that most of the IRS' software is written in COBOL. I can't think of many things more inefficient and cost-ineffective than continually making bit-part upgrades to 60-year-old software written in a language almost no one knows anymore because of a lack of predictable (human and financial) resources needed to rewrite it.

I appreciate your corrections and agree with all of them.

Yet they do not imply IRS is operating efficiently, as efficiency is left undefined.

If we're talking about social efficiency, the tax code appears massively inefficient to me. I believe the tax code should be redesigned to reduce that inefficiency. That may obviate many IRS needs for resources.

> yet they do not imply IRS is operating efficiently

The IRS budget for 2020 was $11.5bln to support $3.42 trillion in tax receipts, so I'm not sure that's inefficient. Could it be more efficient? Sure, but it's not as bad as people like to paint it. For comparison, the Los Angeles Unified School District budget alone for that same year was $15bln.

It doesn't need to be "efficient" in terms of cost, that doesn't really matter. What matters much more is whether it is seen and being effective and fair in collection of taxes & enforcement of violators.

A large part of the Individual Master File processing engine is also written in mainframe assembly language.

But the broader issue seems to be that the IRS is really bad at planning, goal-setting, and managing its development lifecycle [1]. They've burned through $5.9B just on modernization projects since 1998, with very little to show for it. Many of those projects, including replacing IMF and CADE (which was supposed to be the modern system that would replace IMF, but which failed), have been paused indefinitely. Clearly replacing these legacy systems is time-consuming work, but any way you look at it, it should not take 30 years (that's how long it would have taken if they accomplish their current goal of retiring the IMF by 2028).

It's not just that development is slow; this is holding back IRS from operating efficiently. IRS employees reportedly have more than a hundred internal applications they have to be trained in and switch between in order to accomplish basic tasks, and they have a huge backlog.

[1] https://sgp.fas.org/crs/misc/IF12525.pdf

I don't think this issue is specific to the IRS; it has much more to do with how the government is funded - via a series of Continuing Resolutions which make it really difficult to achieve any sort of long-term planning or projects. In the case of BSM funding, your source cites fluctuations are truly wild and were likely enacted on very short timeframes (i.e. months or weeks), which makes it incredibly difficult to hire and retain the expertise or make capital investments you need for success. The debacle around the IRS funding in the Inflation Reduction Act (the subject of the now-deleted parent comment I replied to) highlights these issues - a bunch of funding was allocated, some of it was later clawed back through a budget deal between the White House and Congressional Republicans, and a likely deal this January may continue to reallocate funding. I don't want to lay all the blame for all the IRS' ills at the feet of the state of government funding, but I certainly think it makes it really, really hard to get significant multiyear projects completed.

They did just get a big boost, but this was after years of stagnant funding, as the link in the comment you replied to shows.

Comparing government agencies' efficiency to private entities' is tricky since they tend to have different goals and constraints, but I'm not sure 87,000 agents is that many considering the number of households and business that have to pay taxes, and the number of non-profits that don't pay taxes but still file financial information.

IRS funding is objectively efficient. Every additional $1 spent on IRS budget brings in $5-9 of additional funding through enforcement.

https://www.cbo.gov/publication/57444

No, that just means it makes economic sense to spend it. Doesnt mean that IRS is run the most optimal and efficient way. You can have a very inefficient monopoloy that still makes a lot of money because it is a monopoly. One can imagine where IRS is more effcient and 1 dollar spent results in 20 dollars more etc.

> run the most optimal and efficient way

Why does that matter? No government agency on the planet is run in the most optimal and efficient way; nor businesses or even households. What actually matters is the trend and how it's current efficacy compares to its past.

The efficiency bar has to be particularly low when you consider that it is enforced by law. What incentives do they have to do anything differently when people HAVE to pay them or face consequences.

I'm glad they are at least now trying to have a semblance of customer service awareness.

It's not that easy. The efficiency is actually a benefit for many private companies and it's lobbied very hard. Just look at a comparison with other countries that are also restricted by law and have much nicer systems.

It's very inefficient for the people who are currently stealing those $5-9.

Why don’t they give the irs 100 trillion? America will be rich I say, rich!

The agents being hired are mostly just support agents answering the phone. And on net the investment in improving the IRS is a net positive on revenue because they can more effectively go after folk who are cheating.

It's amazing how some folks will be ardently opposed to "defunding the police" ... except for the tax police.

You do realize why the IRS and cops are different, yes?

it's unpopular because IRS only goes after low-level tax cheats, not the already-rich ones who steal far more, and funding them more doesn't change that priority. similar to police.

In this particular case, this funding does change that priority. Janet Yellen has instructed that "any additional resources—including any new personnel or auditors that are hired—shall not be used to increase the share of small business or households below the $400,000 threshold that are audited relative to historical levels" [1]. Unfortunately, auditing extremely wealthy individuals is complicated and expensive because you're pitting underpaid and overworked IRS employees against a wildly complicated tax system that is exploited to the max by the super-rich and their well-remunerated tax lawyers.

[1] https://home.treasury.gov/news/press-releases/jy0918

https://www.cnn.com/2023/10/20/politics/irs-millionaire-tax-... (“IRS has collected $160 million in back taxes by cracking down on millionaires”)

https://www.irs.gov/newsroom/irs-announces-sweeping-effort-t... (“IRS announces sweeping effort to restore fairness to tax system with Inflation Reduction Act funding; new compliance efforts focused on increasing scrutiny on high-income, partnerships, corporations and promoters abusing tax rules on the books”)

The $80B is new, and until recently they were understaffed and underfunded.

It's like a poor person finally getting Section 8 housing, and then you turn around and accuse them of being wasteful and inefficient with their spending, because they have a house so they clearly aren't as poor as they say.

Also the $80B and 87k agents is slightly overstating what they are actually going to get. See e.g. https://www.forbes.com/sites/insider/2022/10/20/the-reality-....

Is it possible that the IRS lacks the engineering management expertise to build an effective in-house software team? Sure.

Is it possible that they won't spend their money efficiently as a result? Sure.

Is it possible that they'll hire B-grade programmers, because the top people won't want to move to some random suburb in MD to work for $120k/year? Sure.

But to somehow accuse the IRS of having "massive" funding and imply that they are in fact well-supported is just disregarding the facts of recent history.

Tax law enforcement is not the same as the end user experience of taxpayers.

Precisely. They'll spend $80b to hire "agents." They won't spend a time to make the system actually easy to use. They're not underfunded, they're just incompetent.

There's certainly room to debate how the IRS is spending its money, but they've spent some of it on responding to taxpayer questions, resulting in a decrease in wait times and other improvements: https://home.treasury.gov/news/press-releases/jy1421. These are their own metrics, so it's possible they're being gamed in some way, but it's not true no money is going toward helping taxpayers.

Not quite. Of the $79B, $46B was earmarked tax law enforcement, while $4.8B is designated tax processing modernization, including a plan to replace the current legacy systems (IMF and others that are written in a combination of COBOL and assembly language and that run on 1960s mainframes against IBM VSAM data files, and which the IRS have been trying to replace since the 1990s) by 2028. That money is supposed to be spent that way, and the IRS have to get approval from Congress' House and Senate Appropriations Committee (and TIGTA, Office of Management and Budget, and GAO must all sign off on each request) for all their expenses.

I won't disagree that their IT woes seem to be in large part due to incompetence, however.