300,000+ Accountants Left the Profession (2019-2022)—Is This Labor Crisis or Market Correction?

helpful_veteran · April 5, 2026, 4:05am

300,000+ Accountants Left the Profession (2019-2022), and CPAs Are Now Earning Record Premiums—Is This Labor Crisis or Market Correction?

I’ve been following the Bureau of Labor Statistics data pretty closely, and the numbers are staggering: over 300,000 accountants and auditors left the profession between 2019 and 2022—that’s a 17% drop from the peak workforce in 2019.

At the same time, we’re seeing some contradictory signals:

Starting salaries for new accounting graduates jumped ~30% in a single year (2022-2023)
CPA wage premiums are at record levels vs non-CPAs
Unemployment among accounting professionals hovers between 1-2% (nearly everyone who wants work has it)
Spring 2025 undergraduate accounting enrollment jumped 12% year-over-year

So here’s my question: Are we witnessing a profession in crisis, or is this actually a market correction from an oversupplied profession?

The Crisis Narrative

The bears argue this is an existential threat:

300K exodus means talent drain, quality suffers
Workload increases on remaining staff → more burnout → more exits (death spiral)
CPA candidates down 27% over the past decade
Only 1.4% of college students choose accounting (down from 4% a decade ago)
Accounting degrees at a 20-year low (~55,000 awarded vs 124,200 annual job openings)

The data from FloQast showing 99% of accountants experienced burnout and 80% saying monthly close “sabotaged their personal lives” suggests this isn’t sustainable.

The Correction Narrative

The bulls argue the profession was actually oversupplied:

Too many low-skill bookkeepers doing transaction entry (now automated)
Market correcting toward higher-value advisory work
Remaining professionals command premium wages (good for those who stay)
Higher barriers to entry (150-hour CPA requirement) protect wage premiums
Technology eliminates commodity work, increases demand for technical/analytical skills

The Beancount Angle

This is where it gets interesting for our community. Does plain text accounting accelerate or slow this transition?

Acceleration argument: Beancount-style automation eliminates low-skill transaction entry jobs entirely. If you can write a Python importer, you’ve automated what used to be 20 hours/week of bookkeeper labor. This pushes the profession toward higher-value work (advisory, tax strategy, financial planning) where CPAs with licenses command premiums.

Deceleration argument: Beancount makes accounting more accessible to anyone with basic programming skills. You don’t need 150 hours of education or a CPA license to maintain books—you need Git, Python, and willingness to learn double-entry. This reduces barriers to entry and could eventually suppress wages as more tech-savvy non-CPAs enter the market.

My Career Positioning Question

I’ve been mentoring some younger folks lately, and they’re asking: If I’m entering accounting in 2026, should I pursue the CPA license (capture wage premium, professional credibility) or invest in technical skills (Python, data analytics, automation) without the 150-hour education requirement?

The math is brutal. The 150-hour requirement creates what some are calling a “six-figure opportunity cost”—one extra year of school (tuition + lost wages) for a starting salary premium that’s negligible. Meanwhile, technical certifications (AWS, Python, SQL) have much faster ROI.

But the CPA license has regulatory moats (only CPAs can sign audit reports, represent clients before IRS). Those moats protect wage premiums long-term.

Questions for the Community

For professionals here: Are you seeing CPA wage premiums in your market? What’s the differential (CPA vs non-CPA with equivalent experience)?
For solo practitioners/small firms: Is the labor shortage an opportunity (charge more, clients have no alternatives) or a threat (can’t scale, can’t hire staff, growth ceiling)?
For those using Beancount professionally: Do you think plain text accounting skills will eventually command premiums comparable to CPA designation? Or will CPAs who add technical skills command the highest premiums (dual moat)?
For those who left traditional accounting: What made you leave? Was it burnout, compensation, work-life balance, or something else?

I’m genuinely curious whether this 300K exodus is a one-time pandemic-era shock or the beginning of a secular decline. And whether folks like us—using Beancount, learning Python, automating workflows—are positioning ourselves correctly for the profession’s future.

What do you think? Crisis or correction?

accountant_alice · April 5, 2026, 4:06am

This is such an important discussion, and I think the answer is: it’s both—crisis for firms stuck in the old model, correction for those adapting.

I run a small CPA firm in Chicago, and I’ve lived through both sides of this transition. Let me share what I’m seeing on the ground.

The CPA Wage Premium Is Real (For Now)

In my market, the differential is stark:

CPA with 5-7 years experience: $85K-$110K base salary
Non-CPA bookkeeper with equivalent experience: $55K-$70K

That’s a 35-50% premium. But here’s the critical detail: that premium only holds if you’re doing work that requires the license—tax representation, audit sign-offs, attest services.

If you’re just doing bookkeeping, reconciliation, and basic financial reporting? The premium shrinks to maybe 10-15%, and honestly, I’m not sure how long even that will last.

The Hiring Crisis Is Very Real

I’ve had an open position for a senior accountant for 9 months. Nine months! Pre-pandemic, I could fill that role in 6-8 weeks. Now?

Average time-to-fill is nearly 10 weeks (and that’s if you’re lucky)
Each week of vacancy = $3K-$5K in lost productivity
I’ve had to turn down two potential clients because I don’t have capacity

The unemployment rate among accounting professionals being 1-2% means nearly everyone qualified is already employed. You’re not just competing on salary—you’re competing on work-life balance, remote flexibility, meaningful work, career growth.

Where Beancount Fits (And Doesn’t)

Here’s my honest take on the “acceleration vs deceleration” question:

Beancount accelerates the commodity work automation, which is good for the profession long-term. Transaction entry, bank reconciliation, basic categorization—this should be automated. It was never high-value work, and the fact that it consumed 30-40% of junior accountant time was a problem.

BUT—and this is important—Beancount creates a skills gap. Most accountants don’t know Python. Most don’t know Git. Most don’t think in terms of “importers” and “scripts.” So the professionals who can leverage Beancount (or similar plain text tools) have a massive productivity advantage.

I use Beancount for about 30% of my client base now (small businesses who are tech-savvy enough to understand the model). For those clients, I’m probably 40-50% more efficient than I was with QuickBooks. That efficiency lets me:

Charge slightly lower fees (competitive advantage)
Maintain better margins (I’m faster)
Deliver higher quality work (audit trail, version control, reproducible reports)

The Career Advice Question

You asked about CPA license vs technical skills. My advice to anyone entering the field in 2026:

Get both. But prioritize in this order:

Core accounting knowledge (understand debits/credits, GAAP, tax fundamentals)
Technical skills (Python, SQL, data analytics—these have immediate ROI)
CPA license (but only if you want to do work that requires it: audit, tax representation, attest)

The 150-hour requirement is a real barrier, and frankly, I think it’s unsustainable long-term. The profession is going to have to reckon with whether that barrier protects quality or just limits supply to prop up wages.

The Brutal Truth About the Exodus

You mentioned the FloQast study showing 99% burnout. I believe it. I’ve seen it firsthand. Here’s why people left:

Busy season destroyed work-life balance (60-80 hour weeks Jan-April is not sustainable)
Repetitive work felt meaningless (spend 40 hours reconciling bank accounts, get no recognition)
Compensation lagged tech/finance (my friend with similar experience in tech makes 40% more than accounting peers)
Student loan debt vs starting salaries (150-hour requirement = more debt, negligible salary premium initially)
Lack of career growth visibility (“I’ll be doing the same work in 5 years, just with more clients”)

Is This Sustainable?

Short answer: No. The profession has to change. The firms that will survive are the ones that:

Embrace automation (Beancount, AI categorization, RPA for repetitive tasks)
Shift to advisory/strategic work (CFO services, tax planning, financial strategy)
Offer work-life balance (remote work, flexible schedules, no expectation of 80-hour weeks)
Invest in technical training (teach staff Python, SQL, data analytics)
Pay competitively (match tech/finance salaries, or at least get close)

The 300K who left? Many of them went to better-paying, less stressful jobs in tech, finance, or left the workforce entirely. The profession has to compete with those alternatives, not just with other CPA firms.

My Prediction

I think we’re seeing a bifurcation:

Premium tier: CPAs with technical skills, offering high-value advisory work, charging $200-$400/hour, working reasonable hours
Commodity tier: Automated/AI-driven bookkeeping, very low cost, minimal human involvement

The middle tier (traditional bookkeepers doing manual transaction entry for $75/hour) is disappearing. That’s the 300K who left—and they’re not coming back.

For folks in this community using Beancount, you’re positioning yourselves in the premium tier. You’re building technical skills that will be table stakes in 2030. That’s the right bet.

newbie_accountant · April 5, 2026, 4:06am

I’m literally living this transition right now, so this hits close to home.

I’m 30, been in tech (DevOps engineer) for 5 years, and I’m one of those people who chose NOT to go into accounting even though I was interested in it. Here’s why—and why I think the “correction” narrative is more accurate than “crisis.”

Why I (And My Friends) Didn’t Choose Accounting

When I was looking at career options in undergrad (2018), accounting looked like:

5 years of education (4-year degree + extra year for 150-hour requirement)
Starting salary ~$55K (in my market—Portland)
Expectation of brutal busy seasons (everyone I talked to mentioned 70-80 hour weeks Jan-April)
Boring, repetitive work for the first 3-5 years (“paying your dues”)

Meanwhile, tech looked like:

4-year degree (or honestly, boot camp + self-teaching for many of my friends)
Starting salary ~$75K-$85K (same market)
Reasonable hours (40-45/week, flex time, remote options)
Interesting work from day one (building things, solving problems, seeing impact)

The math wasn’t even close. And this was before student loan crisis hit peak intensity.

So I went into tech. And honestly? No regrets on that decision.

But Now I’m Interested in Accounting Again (Because of Beancount)

Here’s the interesting part: I’m teaching myself accounting now, using Beancount, because it’s actually interesting when you approach it like a developer.

Plain text accounting makes sense to me:

It’s version controlled (just like code)
It’s scriptable and automatable (just like infrastructure)
It’s transparent and auditable (just like good systems design)
The data is portable and future-proof (just like using open formats)

If accounting had been taught this way in school—“here’s double-entry bookkeeping, here’s how to model it in code, here’s how to build financial systems”—I probably would have majored in it.

But it wasn’t. It was taught as “memorize journal entries, pass the CPA exam, get a job at Big Four, work 80-hour weeks.”

The “Market Correction” Makes Sense

From my outsider perspective, here’s what I think happened:

The profession was filled with people doing low-skill transaction entry work that should have been automated 20 years ago. QuickBooks, Xero, Wave—these tools should have eliminated most bookkeeping jobs. But they didn’t, because:

The tools weren’t good enough (required lots of manual intervention)
Small businesses didn’t trust automation
Accountants had incentive to keep doing manual work (billable hours)

Then COVID + great resignation happened, and 300K people looked around and said: “Wait, I’m spending 40 hours/week doing data entry that could be automated, I’m getting paid less than my friends in tech, I’m working insane hours during tax season, and I’m supposed to do this for 40 more years? No thanks.”

That’s not a crisis. That’s a rational response to poor working conditions and low wages relative to other options.

The People Who Stayed (And Who’s Coming Back)

The 12% enrollment increase is interesting. I’d bet those are people who:

See the wage premiums and tight labor market (good negotiating position)
Understand technology and want to bring it to accounting
Are interested in the advisory/strategic side (not transaction entry)
Like the stability and regulatory moats of the profession

Basically, people who want to be accountants, not people who couldn’t figure out what else to do.

My Advice to the Profession (From a Tech Outsider)

If you want to attract people like me—tech-savvy, analytical, interested in systems—here’s what you need to do:

Stop treating the first 3-5 years as “paying your dues” with repetitive work. Tech doesn’t do this. Juniors get interesting problems from day one. Yes, they need mentorship and supervision, but they’re not doing data entry for years.
Teach programming and automation as core skills, not “nice to have.” Every accounting grad should know Python, SQL, and Git. Period.
Fix busy season. 80-hour weeks are not sustainable, not healthy, and not necessary. If you need 80-hour weeks to deliver services, your processes are broken.
Emphasize the interesting parts early: financial modeling, tax strategy, business advisory, system design. Not “reconcile 47 bank accounts and we’ll let you do interesting work in 5 years.”
Reduce the 150-hour requirement or eliminate it entirely. It’s a barrier that doesn’t correlate with better outcomes. Make CPA exam harder if you want to maintain quality, but don’t require an extra year of school.

Why I’m Using Beancount (And Not Becoming a CPA)

I’m using Beancount for my personal finances and I’m teaching myself accounting principles because I’m interested in understanding my money, not because I want to work 80-hour weeks during tax season.

But here’s the thing: if there were more jobs that combined accounting knowledge + technical skills + reasonable hours + competitive pay, I would 100% consider transitioning.

That job doesn’t really exist in traditional CPA firms yet. But I bet it will in 10 years. And when it does, you’ll see more people like me entering the profession.

TL;DR

The 300K exodus isn’t a crisis—it’s the market correcting away from low-value, automatable work that was never sustainable. The people left are the ones who want to be there. The profession will be smaller, more technical, and better compensated. That’s a good thing.

And tools like Beancount are accelerating this transition by showing what’s possible when you treat accounting like engineering: reproducible, automated, version-controlled, and intellectually interesting.

bookkeeper_bob · April 5, 2026, 4:06am

I’m one of the 300,000—or at least, I almost was. Let me share my story because I think it illustrates both the crisis AND the correction.

My Near-Exit Story

2021, right in the middle of the exodus. I was working at a mid-sized accounting firm in Austin, doing bookkeeping for ~30 small business clients. My situation:

60-70 hours/week during busy season (Jan-April, then again in Sept-Oct)
$58K salary (no overtime as salaried employee)
Repetitive work: 80% of my time was data entry, bank reconciliation, chasing clients for receipts
No work-life balance: Missed my kid’s soccer games, ate dinner at my desk, constantly stressed

My wife works in tech (project manager at a SaaS company). She was making $95K, working 40 hours/week from home, had flex time, got to see our kids.

I was ready to quit accounting entirely. Started looking at tech boot camps, project management certifications, anything to get out.

What Changed (And Didn’t)

I didn’t leave accounting. But I did leave traditional firms. Started my own bookkeeping practice in 2022, and here’s what’s different:

I set my own hours (no busy season death marches)
I choose my clients (I fire clients who don’t respect boundaries)
I charge what I’m worth ($125/hour, up from effective $29/hour at the firm)
I use modern tools (Beancount, Python scripts, automation)
I work ~35-40 hours/week and make more than I did at the firm

But here’s the key: I can only do this because of the labor shortage. If there were abundant accountants available, clients would just hire someone cheaper. The shortage gives me leverage.

The Crisis Part (It’s Real)

For traditional firms that haven’t adapted, this IS a crisis:

They can’t hire. I have friends still at firms who’ve had open positions for 6-12 months. They’re turning down clients, losing accounts to larger firms with capacity, burning out existing staff with overwork.

They can’t retain. Starting salary jumped 30%, but firms are struggling to match tech salaries. And salary isn’t even the main issue—it’s hours, flexibility, work-life balance. You can’t pay someone enough to work 80-hour weeks when their friend with similar skills works 40 hours remotely for comparable money.

They can’t scale. The traditional model was: hire cheap junior staff, bill them out at 3x their cost, seniors review the work. But you can’t hire cheap junior staff anymore, and juniors don’t want to do repetitive work for 5 years before getting interesting projects.

The Correction Part (Also Real)

Sarah (newbie_accountant) nailed it: the profession was filled with low-value work that should have been automated.

When I was at the firm, I’d spend literally 15-20 hours/week on:

Manually entering transactions from bank statements (should be automated)
Categorizing expenses one by one (AI can do this)
Reconciling accounts (should be 90% automated with 10% human review)
Chasing clients for receipts (workflow problem, not accounting problem)
Copying data between systems (systems integration problem)

None of that is valuable accounting work. It’s data processing that got bundled into “bookkeeping” because someone had to do it.

Now with Beancount + importers + scripts, I’ve automated probably 60% of that work. The time I spend now:

Reviewing categorization (5-10 min/week per client)
Month-end adjustments and accruals (30-45 min/month per client)
Financial analysis and advising (1-2 hours/month per client)
Tax planning conversations (quarterly)

That’s WAY more valuable than what I was doing before. And clients pay more for it because they’re getting strategic advice, not just transaction entry.

Where the 300K Went

Based on my friends who left:

Went to tech: ~40% (better pay, better hours, interesting work)
Went to internal finance/accounting roles: ~30% (corporate accounting, FP&A—better hours than public accounting)
Left workforce entirely: ~15% (mostly women who’d been trying to balance brutal busy seasons with childcare)
Started own practices/consulting: ~10% (like me)
Retired: ~5% (boomers who were retirement-eligible)

The ones who went to tech or FP&A? They’re not coming back. They’re making $80K-$120K with 40-hour weeks. You can’t convince them to return to $65K with 70-hour busy seasons.

The Wage Premium Question

Alice asked about CPA wage premium. Here’s what I’m seeing in Austin:

For traditional firm work: CPA adds maybe 20-30% to salary, but both CPA and non-CPA are overworked and underpaid relative to tech.

For solo practice/consulting: CPA license matters a LOT if you’re doing tax or audit work (regulatory requirement). For bookkeeping/advisory? Clients care more about results, responsiveness, and whether you understand their business. I don’t have CPA, and I’ve never lost a client because of it.

For technical + accounting skills: This is the premium tier. If you can do bookkeeping + Python automation + financial modeling + tax planning, you can charge $150-$250/hour as consultant. CPA on top of that? Maybe gets you to $200-$300/hour.

My Take: Both Crisis and Correction

Crisis for: Traditional firms stuck in old model (hire cheap juniors, bill hourly for repetitive work, expect 80-hour busy seasons)

Correction for: The profession overall (eliminating low-value work, shifting to advisory/strategic, better compensating people who remain)

Opportunity for: People like us who embrace automation, charge for value not hours, and treat accounting like engineering (reproducible processes, version control, continuous improvement)

The Beancount Angle

You asked if plain text accounting accelerates or decelerates this transition. It accelerates it, and that’s a GOOD thing.

Every hour I save on data entry via Beancount importers is an hour I can spend on:

Strategic financial advice
Tax planning
Process improvement
Actually talking to clients about their business

That’s higher value for clients, more interesting for me, and better for the profession long-term.

The 300K who left were doing work that SHOULD be automated. We should be celebrating that they found better opportunities, not trying to drag them back to jobs that shouldn’t exist.

Advice for People Entering Now

If you’re coming into accounting in 2026:

Learn automation from day one (Python, Beancount, whatever—just learn to eliminate repetitive work)
Focus on advisory/strategic skills (tax planning, financial modeling, business strategy)
Set boundaries early (don’t accept 80-hour weeks as “just how it is”—that model is dying)
Get CPA if you want regulatory moat (audit, tax representation), skip it if you’re doing advisory/bookkeeping
Consider solo practice/consulting (labor shortage gives you leverage, you can charge what you’re worth)

The profession is transforming. The people positioning themselves for the future (technical skills + accounting knowledge + value-based pricing) will do very well. The people trying to preserve the old model (hourly billing for repetitive work) will struggle.

I’m glad I didn’t leave. But I’m also glad I left that job. There’s a difference.