The Bureau of Labor Statistics lists on-the-job training for accountants and auditors as None. That is not a comment on how firms behave. It is a classification, and it means the federal profile for the occupation carries no training baseline you could benchmark against. Accounting staff training statistics have to come from somewhere else, and it matters a great deal which somewhere.
Which Training Numbers Are Measured on Accountants
Three sources publish something about training that is specific to this occupation. Anything else is worth testing before you repeat it, because a training percentage pooled across all industries is an economy-wide figure wearing an accounting headline.
| Source | What it records | What it cannot tell you |
|---|---|---|
| The BLS Occupational Outlook Handbook | The occupation's typical entry bar and the training category assigned to it after hire | Hours, or anything about one firm's program |
| State boards of accountancy, and the model rules behind them | Continuing education a licensee must complete to renew | Anything about unlicensed staff, who do much of the preparation work |
| The AICPA Trends report | Graduate supply, and hiring reported by the firms that responded | A population estimate of firm hiring |
Each of those is narrower than the headline it usually gets. Taken together, they also explain why a national average of training hours per accountant has nothing solid to rest on.
What the Federal Occupational Profile Records
Start with the quick facts block at the top of the occupation's profile. For accountants and auditors, Typical Entry-Level Education is a bachelor's degree, Work Experience in a Related Occupation is None, and On-the-job Training is None (BLS, Occupational Outlook Handbook, accountants and auditors).
Read the field definition before you read the value. The handbook defines the on-the-job training field as the "Additional training needed (postemployment) to attain competency in the skills needed in this occupation" (BLS, Occupational Outlook Handbook, accountants and auditors).
So the classification treats someone who clears the entry bar as needing no typical additional training period to reach competency. It is a statement about how the occupation is categorized, not a count of hours, and not a claim that firms train nobody. What it establishes is that there is no federal training baseline for accountants to measure your own program against. The same handbook does assign a training category to bookkeeping, accounting and auditing clerks, one of several differences between those two roles covered in how to hire a bookkeeper.
Continuing Education Is the Only Training Volume Anyone Counts
If you want a training number for accounting staff that somebody records and audits, it is continuing professional education, or CPE. It exists because a license depends on it, and it covers licensees only.
The Model Rule
The National Association of State Boards of Accountancy and the AICPA publish the Uniform Accountancy Act Model Rules, model regulations a state board may adopt into its own rules. Under Rule 6-4, a licensee seeking renewal completes a minimum of 20 CPE credits during each annual period included in the reporting period, an average of no fewer than 40 credits for each annual period within that reporting period, an average of 2 ethics credits for each annual period, and a minimum of 50% of the total credits required for that reporting period in technical fields of study (NASBA and AICPA, Uniform Accountancy Act Model Rules, January 2024).
The rule's own summary chart shows what those minimums add up to. A three-year reporting period requires 120 credits in total, with 20 as the floor in any single year, 6 qualifying ethics credits and 60 credits in technical subject areas (NASBA and AICPA, Uniform Accountancy Act Model Rules, January 2024).
What Your State Board Actually Requires
A model rule binds nobody. Your state board's own version binds, and the boards disagree on the length of the period, the annual floor and the ethics cadence.
California counts across two years. A total of 80 hours of continuing education must be completed in the two-year period immediately preceding license expiration, including a minimum of 20 hours during each year of that period with 12 hours in technical subject matter, four hours of ethics education, and a two-hour Board-approved Regulatory Review course if more than six years have lapsed since the licensee last completed one (California Board of Accountancy, Continuing Education Quick Reference Guide).
New York counts by calendar year and offers a choice. Its CPAs complete either 24 or 40 contact hours of acceptable formal continuing education from January 1st through December 31st of each calendar year, 24 if concentrated in one of the recognized subject areas or 40 if taken in any combination of them, plus 4 contact hours of acceptable professional ethics during every three calendar years (New York State Education Department, mandatory continuing education questions and answers).
A firm with licensees in both states is running two clocks of different lengths against different subject rules. An average across them would describe neither firm nor state.
What a CPE Hour Does Not Tell You
A credit is a unit of sponsored program time, not a unit of learning. What a credit is worth in minutes, and how that changes the real price of a conference, sits in accounting conferences. Completed hours are an attendance figure, which is also what makes them auditable.
The larger limit is coverage. The preparers and bookkeepers carrying most of the volume in a small practice usually hold no license and no mandated floor. Enrolled agents are the exception, and their federal minimum, along with what to write down for everyone else, is set out in training offshore staff.
What the Profession's Pipeline Survey Will and Will Not Support
The AICPA's Trends report is the profession's own count of the accounting pipeline, and its latest edition is unusually direct about the weight its numbers can carry.
Responding firms reported hiring a total of 11,985 new graduates in 2024, and 75% of those, or 8,994, were accounting graduates (AICPA, 2025 Trends: A Report on Accounting Education, the CPA Exam, and Public Accounting Firms' Hiring of Recent Graduates). The rest arrived holding other degrees, and that share is a training fact before it is a hiring fact.
The methods appendix is where the caveat lives. On the supply side, a census of 1,174 colleges and universities granting accounting degrees was invited, and the response rate was 4%, or 49 institutions. On the demand side, a census of 19,097 public accounting firms was invited, and the appendix records the response rate as 01.00% (AICPA, 2025 Trends report).
The report draws the conclusion itself rather than leaving it to the reader. Response volume was "insufficient to project hiring trends for 2024 graduates (across bachelor's, master's, and nonaccounting degrees) with statistical reliability", and the new graduate hiring total "could not be projected with confidence" (AICPA, 2025 Trends report).
That does not make the hiring total useless. It makes it a floor from the firms that answered rather than an estimate of the profession, which matters the moment somebody converts it into a rate. The supply-side count of degrees awarded, the figure quoted most often from this report, is in how to hire a staff accountant.
Four Questions That Test Any Training Statistic
A figure that cannot answer the first two of these is not a benchmark, however it is being used.
- Who was asked? A survey of employers across all industries reports one pooled percentage unless it breaks its results out by sector, so on its own it describes an economy rather than a practice.
- How many answered? A published response rate is what separates a sample from a mailing list. Where the rate is missing, assume the respondents chose themselves, since an employer with a program worth describing is likelier to fill in a training survey.
- What period does it cover? Annual, biennial and triennial reporting are three different measures, and stacking them into one figure produces arithmetic rather than information.
- What is the denominator? Hours per employee, hours per learner, and hours per person who completed at least one course are three different numbers from the same raw data, and the one in play often goes unnamed.
None of this means a figure that fails the test is false. It means the figure cannot be applied to your firm, which is a quieter problem and an easier one to miss.
What Accounting Research Measures When It Studies Training
The peer-reviewed accounting literature does study training, and reading one recent study closely shows how far that evidence sits from a working firm.
A study in The Accounting Review tested whether telling people how they rank against their peers, which it calls relative performance information, changes whether they take training that is offered to them. Relative performance information "increases the likelihood of taking training when a work-training tradeoff is absent but decreases it when a work-training tradeoff is present", and it raises performance when that tradeoff is absent while the effect is less positive when it is present (Arnold, Shi, Tafkov and Voermans, The Effects of Relative Performance Information and Work-Training Tradeoff on Employees' Skill Development, The Accounting Review).
Now read the method. Participants solved multiplication problems over seven rounds, and the researchers "recruited 90 undergraduate students from a large U.S. university" for the initial data collection plus 50 more for an additional data collection (Arnold, Shi, Tafkov and Voermans, The Accounting Review).
The mechanism is real and carefully tested, and the tradeoff it manipulates is the same kind your firm creates whenever training is scheduled inside chargeable time. It is still not a measurement of accountants learning accounting work, and that gap is the honest state of the evidence rather than a criticism of the study.
Questions Firms Ask About Accounting Staff Training Statistics
What Is the 70-20-10 Rule for Employee Development?
It is a guideline for developing managers, and its publisher says so plainly. The Center for Creative Leadership describes the ratio as 70% challenging experiences and assignments, 20% developmental relationships and 10% coursework and training, and says it emerged from over 30 years of its Lessons of Experience research into how executives learn, grow and change across a career (Center for Creative Leadership, The 70-20-10 Rule for Leadership Development).
As a design prompt it earns its keep, because it asks whether your staff are given the stretching assignments the largest share of the ratio assumes. As a statistic about accounting staff it does no work at all, since firm records rarely split one preparer's learning into those three shares.
What Should an Accounting Firm Track Instead?
Numbers you can compute from your own records, with the denominator written down before you start. Three are worth the effort.
Track continuing education completion against each licensee's own state floor rather than against a national figure, since the floor is the only requirement anybody can enforce. Track the share of incoming staff arriving without an accounting degree, because that is training load you can predict before a start date.
Third, track whether your own training hours have a stated denominator at all. A number that changes meaning between two conversations is not yet a metric. Wider firm measures sit in accounting firm KPIs.
Is There a Reliable Average of Training Hours per Accounting Employee?
No, not from the sources that measure this occupation, and none of them clears the four questions above. The federal occupational profile records a category rather than hours, state continuing education floors are counted over reporting periods of different lengths, and the profession's own survey declines to project its demand-side numbers. An average built across those is an average of things that are not the same measure.
Count Your Own, and Write Down the Denominator
The useful move is to stop hunting for a benchmark that does not exist and start recording a number that does. Your licensees have a floor set by their own board. Your incoming staff have degrees you already know about. Your training hours have a denominator, whether or not anyone has written it down, and writing it down is what turns them into something you can compare next year.
If you are counting training load because you are deciding whether to add capacity, don't trust us, test us. Accountably places trained offshore accountants and tax preparers inside US CPA and EA firms, ramped on your software and SOPs in about 3 to 4 weeks, and since 2022 that has meant 30+ placements across 20+ US firms. The low-risk way in is a Free 40-Hour Proof Pilot, a fixed block of your own representative work prepared on your SOPs and put through full multi-layer review, so your reviewer grades real output before a client file is at stake. If a placement is not a fit in the first 30 days, we replace them free.