Inside the Minds of Taxpayers Who Owe

Executive Summary

Enforcement Got Quieter. Belief in It Got Quieter Too. The Notices Didn't.

Optima's own client base, federal enforcement data, and independent government audits all point the same way: the IRS that taxpayers picture (agents, audits, liens) has been retreating for years. What replaced it is smaller, more automated, and, in one specific way, more effective than most taxpayers realize.

51%

Don't believe the IRS enforces against people who owe, or aren't sure

18%

Say they'd actually trust the IRS to help them

88%

Couldn't confidently say what to do about their tax problem

+271%

More taxpayers respond once an automated IRS notice reaches them

−30%

IRS workforce cut between Jan. 2025 and Jan. 2026

The Central Finding

The enforcement Americans can see has been retreating for years. The enforcement they can't see just picked up the pace.

Since January 2025, the IRS has lost roughly 30% of its workforce. Collection staff are down 41% since FY2024. Appeals revenue has fallen 71% in three years. Inflation Reduction Act enforcement funding ran out completely on December 31, 2025. None of this is a single bad year. It's the tail end of a multi-year retreat.

Taxpayers have registered the change, and their read is largely fair: just 18% say they'd trust the IRS to help them resolve a problem, and 51% either don't believe the agency enforces against most people who owe, or aren't sure it does.

But one program tells a different story. The IRS's Automated Substitute for Return process, which issues a computer-generated assessment whenever a return never arrives, closed 592,800 cases in FY2025, up 135% in two years. Taxpayer engagement with those notices is climbing even faster than the notices themselves, up 271%. The machine didn't stop. It just stopped introducing itself with an agent, a lien, or a knock at the door.

The Retreat

A Retreat, Not a Blip

The pullback in enforcement capacity isn't new, and it isn't over. Taxpayers are living through the back half of a multi-year drawdown, against a problem that keeps growing.

The Problem Is Bigger Than It Used to Be

The federal Tax Gap: tax legally owed versus tax actually paid on time, Tax Year 2022

$696B

Total gross Tax Gap, TY2022

14.7M

Potential nonfilers identified, TY2022, up 67% since TY2015

95%

Of that growth came from earners under $200,000

$539B

Underreporting (income understated on filed returns)

$94B

Underpayment (filed but didn't pay)

$63B

Nonfiling (never filed at all)

Source: IRS Tax Gap Projections, Pub. 5869; TIGTA nonfiler-coordination audit, 2026

Fewer People, Doing More

IRS enforcement staffing, January 2025 to January 2026

−46%

ACS tax examiners lost (1,842 positions)

−40%

Field Collection employees lost (1,429)

−37%

CSCO tax examiners lost (873)

103K→72K

Total IRS employees, Jan 2025 to Jan 2026 (31,273 separations)

Source: TIGTA workforce and compliance-trend audits, 2026

Where the Enforcement Dollars Come From Now

IRS enforcement revenue, FY2023 through FY2025

87%

Share of all enforcement revenue from Collection in FY2025, up from 81%

−71%

Appeals revenue, three-year decline

−35%

Examination revenue, year over year in FY2025

+9%

Total enforcement revenue, three-year change

Source: TIGTA compliance-trend audit, FY2025

The Money Behind the Machine Is Gone

98%

Of $3.8B Inflation Reduction Act enforcement funding spent, fully exhausted Dec 31, 2025

−29%

Projected FY2026 enforcement spending vs. FY2025

TIGTA Finding: Cases Stuck in the System

Nearly 40,000 high-priority cases sat frozen for two years, and nobody outside the IRS knew.

The IRS's high-income nonfiler initiative, launched in February 2024, tracks 135,270 cases involving taxpayers earning $400,000 or more, plus federal and IRS employees, making it one of the agency's most closely monitored enforcement efforts. Even so, as of June 2025, TIGTA found 38,824 of those cases covering 33,653 taxpayers, an estimated $15.7 billion in potential assessments, had received a first notice in early 2024 and then gone silent. A technology hold meant to be temporary froze the entire sequence. The IRS didn't move them forward until March 2026.

Another 10,969 cases (8,853 taxpayers) sat unworked in the queue awaiting assignment. And 4,918 cases (4,748 taxpayers) got delinquency notices despite having already filed; their returns simply hadn't posted yet, in some cases for over a year.

Why it matters: if the IRS's own top-priority initiative can lose track of cases for two years, the everyday backlog is unlikely to be moving faster. Penalties and interest kept accruing the entire time, and taxpayers flagged for unfiled returns can be locked out of installment agreements and offers in compromise, the exact relief they need most.

The Trust Gap

Why Fewer Americans Believe Anyone Is Watching

Ask taxpayers directly, and the skepticism isn't a vague feeling. It's a majority position.

Do You Believe the IRS Finds and Enforces Against Most People Who Owe?

51%

Say no, or aren't sure, that it does

49%

Say yes, it does

The full breakdown

49%

Yes

11%

No

39%

Not sure

“I trust the IRS would help me resolve my tax situation if I reached out.”

52%

Disagree

31%

Neutral

18%

Agree

32%

Completely Disagree

31%

Neutral

20%

Slightly Disagree

10%

Slightly Agree

8%

Completely Agree

The Machine

The Machine Nobody Sees: Why Notices Still Work

While the human side of enforcement recedes, one automated program keeps closing more cases and getting more responses every year, without a single knock on the door.

The Collection Notice Funnel, FY2025

Automated notices issued along the IRS's collection escalation path

~13.3M

CP14: Initial balance due notice

~12.3M

CP521: Installment agreement reminder

~8.5M

CP504: Final notice, intent to levy

~5.8M

CP503: Second reminder

~4.2M

CP501: First reminder

~2.7M

CP523: Installment agreement default

339K

Notice of levy issued

214K+

Notice of Federal Tax Lien filed

Source: CP-series notice volumes, IRS notice issuance data, FY2025 annualized. Lien and levy totals: IRS Data Book, Table 4-1, FY2025.

Liens and levies remain the exception, not the rule. The IRS sends millions of "you owe" and "final warning" notices a year, roughly 13.3 million CP14s and 8.5 million CP504s. Per the IRS Data Book, FY2025 collection actually escalated to a lien on just over 214,000 accounts (up 9% from FY2024, up 36% since FY2022) and a levy on 339,137 more (up 8% from 313,792). Both ticked up year over year, and both are still a small fraction of the notices sent. For debt above $65,000, there's a fourth lever the Data Book doesn't even list: the IRS certified 148,508 taxpayers as "seriously delinquent" in FY2025, a step that puts the State Department on notice to deny or revoke a passport, and reversed 106,292 such certifications that same year, though only for debts actually resolved (paid in full, discharged, or settled), not merely paid down. Millions of taxpayers are holding a letter that threatens something that, statistically, rarely follows. That gap between threat and follow-through, at every level from a lien to a passport, goes a long way toward explaining why so many say they're "not sure" the IRS follows through.

The Nonfiler Notice Restart

Individual nonfiler notices issued, by fiscal year

0

FY2023

1.68M

FY2024

3.16M

FY2025

Source: TIGTA nonfiler-coordination audit, 2026

Automated Substitute for Return (ASFR)

When the IRS files a return on your behalf, using only what third parties reported

592,800

Total ASFR cases closed, FY2025 (+135% vs. FY2023)

430,900

Closed without any taxpayer contact

161,900

Closed with taxpayer contact (+271%)

$2.9B

Additional assessments, FY2025

Source: TIGTA compliance-trend audit; IRS Data Books, FY2023–FY2025

Perception vs. Reality: What's Already on the Record

Enforcement Action	Self-Reported	On Record	Gap
IRS letters escalated over time	54%	~58%	aligned
Substitute for Return filed	2%	~12%	+10 pts
Federal tax lien	6%	~8%	+2 pts
Levy / garnishment activity	10%	~26%	+16 pts
No enforcement action at all	29%	~15%	−14 pts

Source: Optima survey self-reports compared with de-identified account records

The Decision

The Decision to Finally Act

When these taxpayers do reach out for help, the data describes a calculated decision, not a panicked one.

Did They Know? Did They Know What to Do About It?

And did they clearly understand what steps they needed to take?

12%

Yes, knew what to do

35%

Somewhat, but not confident

53%

No, did not understand options

Before contacting Optima, were you aware you were out of compliance with the IRS?

50%

Yes, and I delayed dealing with it

18%

Yes, but I underestimated the consequences

17%

I suspected it but wasn't sure

15%

No, I did not realize I was noncompliant

What Kept Them From Acting Sooner?

29%

I didn't believe I could afford any solution

23%

I felt overwhelmed and avoided it

23%

I didn't know where to start

12%

Afraid contacting the IRS would make it worse

10%

Didn't understand what the IRS would allow

3%

Hoped it would resolve itself

What Ultimately Triggered Action?

Select up to two, share of total selections

29%

I realized waiting was making it worse

21%

Fear of garnishment or levy

19%

The balance grew beyond what I could ignore

15%

An IRS enforcement notice or deadline

12%

A major life event (job loss, illness, divorce, etc.)

4%

Someone urged me to get help

Did You Try Resolving It Directly With the IRS First?

30%

Yes

70%

No

Why Did You Seek Professional Help Instead of Handling It Yourself?

38% vs. 5%

Sought help chasing a better outcome, not out of fear of IRS enforcement

38%

I wanted the best possible outcome

24%

I didn't trust myself to handle the IRS correctly

14%

IRS rules and notices too confusing

12%

Wanted someone else to deal with the IRS

5%

Afraid of IRS enforcement

5%

Other

How Did It Feel to Finally Reach Out?

36%

Overwhelmed

33%

Ready to fix it

15%

Concerned

11%

Panicked

5%

Defeated

What Would Have Helped Them Avoid This in the First Place?

32%

Affordable early-stage help

18%

Nothing, my situation was unavoidable

16%

Knowing filing is required even if you can't pay

15%

Better tools to estimate what I would owe

12%

Clearer IRS explanations in plain English

8%

Earlier warnings before enforcement

The Data

Who We Heard From

Optima surveyed everyone who enrolled as a client during 2025, not a sample of them, to understand taxpayer behavior and sentiment as directly as possible.

What follows pairs that outreach with de-identified account records for the same population: not who they are demographically, but what their compliance history actually looks like.

The Balance They're Carrying

~3.6

Average unfiled tax years

~$19K

Median total accrued balance

~$38K

Average total accrued balance

~30%

Carry six or more unfiled years

~13%

Zero unfiled years

~31%

1–2 unfiled years

~26%

3–5 unfiled years

~24%

6–9 unfiled years

~6%

10 or more unfiled years

Source: Aggregated, de-identified client compliance records

What's Already on Their Record

Share of clients with each enforcement action on file

~58%

IRS collection letters issued

~26%

Levy action recorded

~20%

Final notice issued

~12%

Substitute for Return filed by the IRS

~10%

Referred to third-party collection

~10%

At risk of passport restriction

~8%

Federal Tax Lien filed

~3%

Passport Restricted

Source: Aggregated, de-identified client compliance records

How Long They've Been Carrying It

~5.8 yrs

Average time since first balance-due year

~5 yrs

Median time since first balance-due year

2020

Median earliest balance-due tax year

~17%

Within last 2 years

~30%

3–5 years

~28%

6–8 years

~21%

9–12 years

~4%

13 years or more

Source: Aggregated, de-identified client compliance records, earliest balance-due filing date

The Bottom Line

Three data sources. One shape.

The retreat is real. IRS capacity is down roughly 30% in a year, Collection staff down 41% since FY2024, Appeals revenue down 71% in three years, and Inflation Reduction Act enforcement funding is gone.

The distrust is real, and defensible. Just 18% trust the IRS to help them. 51% don't believe it enforces against most people who owe, or aren't sure. The government's own auditors keep finding the kind of dysfunction that justifies the doubt.

But the notices are working. ASFR closures are up 135%. Taxpayer engagement with those notices is up 271%. $2.9 billion in fresh assessments came out of a program most people don't even know exists.

And the people who ask for help aren't the panicked ones. They're the ones who ran the numbers and decided expertise beats guessing alone.

The gap that matters isn't between what the IRS used to do and what it does now. It's between what the IRS is actually doing and what taxpayers think it's doing. Closing that gap, one taxpayer at a time, is the work ahead.

Methodology & Sources

Survey. Optima invited every client who enrolled during 2025, the full enrollment population rather than a sample, to share their experience anonymously and help us better understand taxpayer behavior and sentiment. Responses were collected between May and June 2026. Multi-select questions are reported as a share of total selections rather than share of respondents. All survey data is self-reported and subject to response and recall bias.

Client compliance records. Aggregate, de-identified account data for the same 2025 enrollment population. All figures reported are cohort-level averages, medians, or distributions; no individual record is represented or reconstructable. Field definitions may differ slightly from the phrasing used in corresponding survey questions, which may account for part of the perception-versus-record gaps described above.

Federal data. IRS Fiscal Year 2025 Data Book; IRS Publication 5869, Federal Tax Compliance Research: Tax Gap Projections for Tax Year 2022; IRS notice issuance data, FY 2025; IRS Filing Season Statistics. TIGTA audit reports on nonfiler coordination, compliance trends through FY2025, Inflation Reduction Act spending, and IRS workforce levels, published August–September 2026.

Limitations. This report describes a specific population of taxpayers who sought professional tax resolution assistance and should not be read as representative of the general U.S. taxpayer population. Nothing in this report constitutes tax, legal, or financial advice. Individual outcomes vary; published IRS acceptance and resolution rates are national aggregates and are not predictive of any specific case.

© 2026 Optima Tax Relief, LLC · 6 Hutton Centre Dr., Suite 300, Santa Ana, CA 92707 Optima Tax Relief is a tax resolution firm and is not affiliated with the Internal Revenue Service. This report is provided for informational purposes and does not constitute tax, legal, or financial advice.

Tax Relief Services

Optima Tax Shield