Elon’s report card

[jon-nyc]

@Horace said in Elon’s report card:

Why haven’t we?

Diminishing returns and unpopularity as you get to more mom and pop type audits. Political slogans such as ‘an auditor in every kitchen’, etc

I’ve never trusted those stats personally.

It seems obvious to me that a few 150k per year accountants could audit Exxon, Oracle, Trump Enterprises, etc etc and pay for themselves, probably true in average for the entire Russel 2000. And the vast majority of privately held businesses with, I don’t know, 250MM revenue or more?

Probably knowing that there’s some loss-leader audits on smaller peeps keep enough people honest so there’s some payoff there too.

[Horace]

@Axtremus said in Elon’s report card:

Why should we to begin with? What's the rationale for using "until they don’t pay for themselves" as the stopping condition?

Because their job is to collect legally owed tax money, and the government is entitled to all legally owed tax money, by definition. If the government can spend $1 of tax money to get $2 of tax money, that is in the public interest. It's in the interest of those who pay their fair share of taxes.

You're not going to argue to use "until they start paying for themselves" as the stopping condition for firing any particular class of government employees, are you?

Let's say I did, what would you do about it? Write another nonsensical response? Is that the price I would pay for having that opinion?

[Horace]

@jon-nyc said in Elon’s report card:

@Horace said in Elon’s report card:

Why haven’t we?

Diminishing returns and unpopularity as you get to more mom and pop type audits. Political slogans such as ‘an auditor in every kitchen’, etc

I’ve never trusted those stats personally.

It seems obvious to me that a few 150k per year accountants could audit Exxon, Oracle, Trump Enterprises, etc etc and pay for themselves, probably true in average for the entire Russel 2000. And the vast majority of privately held businesses with, I don’t know, 250MM revenue or more?

Probably some loss-leaders on smaller peeps keep enough people honest so there’s some payoff there too.

I understand there can be vague and unprovable reasons for believing that pre-Trump, we were at the "right amount" of auditors, hitting a sweet spot between political viability and revenue maximization, but I remain skeptical that we had this easy financial win at the IRS by just hiring more 150k/year accountants to audit big business.

[jon-nyc]

That’s a nice strawman but no one here has claimed that anyone could find the optimal solution, let alone that we had achieved it.

This issue has been pretty well studied and confirmed with natural experiments.

[jon-nyc]

What data do we have to show that adding auditors at IRS leads to a net increase in government revenues?

There is substantial empirical evidence that adding auditors at the IRS leads to a net increase in government revenues—often significantly more than the cost of hiring them. The core idea is simple: more audits deter tax evasion and recover unpaid taxes.

Here are some key findings from prominent studies and government reports:

⸻

1. Congressional Budget Office (CBO) Estimates
   • CBO (2022) projected that allocating $80 billion over 10 years to the IRS (as done in the Inflation Reduction Act) would increase federal revenues by $180 billion, a return of more than 2:1.
   • The revenue comes from both direct enforcement (e.g. audit findings) and indirect effects (voluntary compliance due to fear of audits).

⸻

2. Treasury Department Estimates
   • The U.S. Treasury (2021) estimated that each additional dollar spent on IRS enforcement yields $5–$9 in increased revenue.
   • Treasury projected that restoring audit rates to early 2010s levels could bring in $700 billion over a decade.

⸻

3. Academic Studies
   • Guyton, Langetieg, Reck, Risch, Zwick (2021, updated 2023): Using randomized audit data, they estimate the return on investment (ROI) of audits is 2.6x on average, but much higher for high-income individuals, sometimes exceeding 10:1.
   • They also found indirect effects: people reduce evasion for years after being audited, which boosts future compliance.

⸻

4. IRS Operational Data
   • The IRS itself has long tracked that enforcement revenue per dollar spent varies by target group but generally ranges from $4 to $8 per $1 spent.
   • However, the return is not linear: returns decline at the margins, especially if auditors are spread too thin or poorly trained.

⸻

5. Government Accountability Office (GAO)
   • The GAO has repeatedly reported that reductions in audit rates, especially for high-income earners and large corporations, have coincided with increased tax gap estimates (now roughly $540–600 billion/year).

⸻

In short: Decades of evidence from the CBO, Treasury, IRS, and academic economists consistently show that investing in IRS auditing capacity leads to net increases in revenue—often at a very high return on investment. The key challenge is deploying resources efficiently (targeting noncompliance effectively) and maintaining audit quality.

Would you like a summary chart or links to primary sources for any of these findings?

Is there any data showing otherwise?

There is little to no empirical data showing that increasing IRS auditors fails to produce a net revenue gain—and virtually no credible evidence that it results in a net loss. However, a few caveats and critiques are worth considering:

⸻

1. Diminishing Returns at the Margins
   • Some studies note that returns are not constant. As the IRS audits easier cases or saturates high-return targets, the marginal return on each additional auditor can decline.
   • Example: Auditing simple wage-earners tends to yield far less per audit than auditing high-income individuals or large corporations.

⸻

2. Poor Implementation Risk
   • If enforcement funding is poorly allocated—e.g., toward unproductive audits or inadequate training—the ROI could be lower.
   • For example, some GAO reports have warned about inefficient use of resources or outdated technology undermining audit effectiveness.

⸻

3. Short-Term vs Long-Term Revenue
   • Enforcement may generate less revenue in the very short term than projected if:
   • Legal challenges delay collections.
   • Audits focus on deterrence rather than immediate collection.

⸻

4. Political/Ideological Critiques (not empirical)
   • Some critics (e.g., in Congress or think tanks like the Heritage Foundation) argue that increased IRS enforcement might:
   • Burden small businesses or lead to overreach, potentially chilling legitimate activity.
   • Lead to costly litigation, offsetting some gains.
   • But these critiques are largely anecdotal or philosophical and not supported by quantitative studies showing net revenue loss.

⸻

5. Historical Budget Constraints Can Mask ROI
   • The IRS budget was cut by ~20% from 2010–2020, and audit rates fell—particularly for high earners. This led to reduced enforcement revenue, but the counterfactual (what revenue would have been with more auditors) is not directly observable, which critics sometimes use to argue against spending increases.
   • However, studies using modeling or natural experiments (e.g. randomized audits) consistently find that more enforcement would have increased revenue.

⸻

In Conclusion

There is no solid data showing that adding auditors results in a net loss of revenue. The most nuanced findings suggest that returns can vary based on how funds are deployed, but even conservative estimates show a positive return. The main debates are about efficiency, fairness, and targeting—not whether the basic ROI is positive.

Would you like to see a table comparing ROI estimates across different income groups or audit strategies?

[Horace]

@jon-nyc said in Elon’s report card:

That’s a nice strawman but no one here has claimed that anyone could find the optimal solution, let alone that we had achieved it.

No, you only implied it by giving reasons for the status quo, constraining it on both ends. Your implications are yours, and they are not a strawman. They are your direct implications.

This issue has been pretty well studied and confirmed with natural experiments.

Ok then, it's just an obviously stupid thing that we don't hire more IRS auditors. Which is fine. I understand that that is the implication of the statistics as they exist. I imagine there are actual reasons we don't hire more auditors beyond political messaging, which is an insubstantial argument at best. it would not be difficult to message that the wealthy are not paying their fair share, or large corporations. That's actually a huge intersection of the venns of both parties.

[jon-nyc]

@Horace said in Elon’s report card:

@jon-nyc said in Elon’s report card:

That’s a nice strawman but no one here has claimed that anyone could find the optimal solution, let alone that we had achieved it.

No, you only implied it by giving reasons for the status quo, constraining it on both ends. Your implications are yours, and they are not a strawman. They are your direct implications.

Oh come on, the only implication of my post was we were on the left side of the optimal point ergo returns are still upward sloping. I would speculate we always have been and that the optimal point, if it could be achieved, would be politically untenable.

[Horace]

@jon-nyc said in Elon’s report card:

@Horace said in Elon’s report card:

@jon-nyc said in Elon’s report card:

That’s a nice strawman but no one here has claimed that anyone could find the optimal solution, let alone that we had achieved it.

No, you only implied it by giving reasons for the status quo, constraining it on both ends. Your implications are yours, and they are not a strawman. They are your direct implications.

Oh come on, the only implication of my post was we were on the left side of the optimal point ergo returns are still upward sloping.

Which was the point of my post, that if that's true, then it's just stupid to not hire more.

Which you argued against by hand waving something about political messaging. You wanted me to be wrong and right at the same time.

So we can agree that I was right that it's just stupid to not hire more IRS agents, if those statistics can be taken at face value, and we can discard your argument about political messaging as a non-point meant to hand-wave the implication that we were actually at a good spot with the status quo.

I would speculate we always have been and that the optimal point, if it could be achieved, would be politically untenable.

Of course it would be perfectly tenable. "We hire more tax auditors to get the lawbreakers to pay their fair share, we get this much more in taxes, your tax burden as a lawful tax payer is this much less". That is not non-starter territory as political messaging goes. Dems want to raise lawful taxes, but you think nobody will get elected by enforcing existing taxes? "We can either increase head count at the IRS, or raise taxes, for the same result, which do you prefer, voter?"

[jon-nyc]

Is there any evidence that we’ve ever had so many auditors that we had gone beyond the point where returns were positive?

There is no evidence that the IRS has ever hired so many auditors that it crossed the point of diminishing returns to the extent that returns became negative or even net neutral. All available data—from historical audit rates, enforcement spending, and revenue returns—suggests that the IRS has long been operating well below the level of maximum return.

Key points:

⸻

1. Historical Audit Trends Show Undershooting, Not Overshooting
   • Audit rates peaked in the 1960s–1980s but were still only around 1–2% of individual returns, compared to ~0.4% today.
   • Even during those higher-audit eras, enforcement revenue more than paid for itself, and there’s no indication the IRS over-audited to the point of inefficiency.

⸻

2. CBO and Treasury Projections Assume Diminishing Returns — and Still Show High ROI
   • Models from the CBO, Treasury, and academic economists already account for diminishing marginal returns, yet they continue to estimate ROI in the 2:1 to 10:1 range.
   • That suggests we are nowhere near the point where returns flatten out completely, let alone turn negative.

⸻

3. Post-2010 IRS Cuts Reduced Audit Rates — and Revenue
   • The IRS enforcement workforce shrank ~30% from 2010 to 2020.
   • Over that period, the tax gap grew significantly, particularly among high-income filers.
   • If the IRS had previously been overstaffed, cutting auditors would have improved efficiency or held revenues steady—but the opposite happened.

⸻

4. Academic Studies Suggest the Real “ROI Frontier” Is Far Ahead
   • Guyton et al. (2021) found that auditing high-income earners yields extremely high returns, and the IRS currently audits them at historically low rates.
   • This suggests that expanding targeted audits could still produce high marginal returns before hitting any true “ceiling.”

⸻

5. No Published Evidence of Net Revenue Decline from Over-Auditing
   • Despite ideological claims of IRS “overreach,” no peer-reviewed study or government report has documented a revenue drop attributable to excessive auditing.

⸻

Bottom Line:

There’s no historical or contemporary evidence that the IRS has ever reached an enforcement staffing level where returns turned negative. On the contrary, the evidence overwhelmingly suggests that the agency is far below optimal staffing, especially for auditing complex, high-yield cases.

Would you like a graph of historical audit rates vs. enforcement revenue?

[Horace]

So it's just an obviously stupid thing to not hire more auditors. I never claimed flat out that it wasn't. I'm aware of obviously stupid things existing. I was only registering my skepticism. I am aware that the statistics at face value have this implication of an obviously stupid thing happening, though I've heard few point out the obvious question of "why not hire more".

[jon-nyc]

It’s not politically teneble.

We’d reduce speeding and save lives if we put traffic cameras literally everywhere or mandated that car manufactures track speeds and report violations to the police.

Good luck to the politician that proposes that. Though the corrupt officials in Yonkers seem to have come close with the speed cameras

[Horace]

@jon-nyc said in Elon’s report card:

It’s not politically teneble.

You can keep saying that, it will remain untrue. There is nothing untenable about messaging an increase in tax revenues, without increasing taxes. Actually, that would be good messaging.

We’d reduce speeding and save lives if we put traffic cameras literally everywhere or mandated that car manufactures track speeds and report violations to the police.

Good luck to the politician that proposes that. Though the corrupt officials in Yonkers seem to have come close with the speed cameras

It's not as if an increase in IRS head count needs to be part of any politician's platform. It would just be buried in some bill, and the wonks would talk about it on podcasts.

Thinking about it more, I suspect that the real reason this isn't done, is that the returns are trivial, if positive.

[jon-nyc]

I agree there exists a possible world where all the evidence is wrong and you’ve intuited the true answer. I just don’t think we live in that world.

[Horace]

@jon-nyc said in Elon’s report card:

I agree there exists a possible world where all the evidence is wrong and you’ve intuited the true answer. I just don’t think we live in that world.

You keep hoping I'm making points other than the ones I'm making. Sorry to not leave the dunking lane open as wide as you'd like.

[jon-nyc]

Ok, Horace. People can read the thread and come to their own conclusions.

[Horace]

@jon-nyc said in Elon’s report card:

Ok, Horace. People can read the thread and come to their own conclusions.

Maybe a small handful of people will read and come to their own conclusions. More likely, nobody will read. (Especially not your AI dumps.) Meanwhile, you would be unable to put a point I've made into your own words and demonstrate where it's wrong. Skepticism is not a factual claim. I also posited this, which ties everything up into a neat bow, but which is just a guess:

Thinking about it more, I suspect that the real reason this isn't done, is that the returns are trivial, if positive.

So go ask Grok what the actual returns are. How much revenue are we going to lose by losing those auditors due to DOGE, how much would be stand to gain by rehiring them and then hiring some more?

[taiwan_girl]

I read this thread. LOL

[Horace]

@taiwan_girl said in Elon’s report card:

I read this thread. LOL

I would never have recommended that.

[LuFins Dad]

[Horace]