In option-pricing and complex-securities valuation, volatility governs the dispersion of future equity outcomes around today’s value, and because option-theoretic instruments have convex payoffs, even modest changes in the assumption can move conclusions materially. That sensitivity is why the phrase “volatility haircut” creates tension. Sometimes a downward adjustment is economically sensible. Peers may be larger, less levered, more diversified, or measured over a different horizon. Sometimes it isn’t, and the difference is whether the report can explain why this company, this security, and this date justify a number that diverges from the unadjusted evidence.
The credibility question turns less on whether the analyst used the word “haircut” than on whether the choice can be defended in the language reviewers actually use: peer-set construction, term alignment, leverage normalization, calibration discipline, and consistency from one valuation date to the next. None of that is new. What’s changed is how systematically auditors, the SEC staff, and the AICPA working draft now look at it.
What the SEC actually requires
- Companies may begin with historical volatility, but must consider how expected volatility may differ.
- Implied volatility is useful; selection requires good-faith effort to use sufficient information.
- For nonpublic and newly public entities, peer selection considers industry, stage, size, and leverage.
- An industry index may help identify peers, but the index’s own volatility is not the company’s input.
- Historical lookback should generally be commensurate with the expected or contractual term.
- Applies to share-based awards granted while the issuer holds positive MNPI, including spring-loaded awards.
- Companies should consider whether to adjust observable market price or expected volatility.
- A facts-and-circumstances reminder, not a license for arbitrary volatility management.
- If important information sits inside the company but isn’t reflected in market prices, inputs may need to move toward the facts.
- This is the SAB most likely to surface in audit and SEC review of pre-event grants.
What the AICPA draft adds
The updated AICPA working draft, released for download in January 2026 from a December 18, 2025 PDF, establishes a fact-specific framework rather than a numerical haircut, and that framework sometimes points the other direction. In the volatility appendix, the draft says private-company volatility is typically based on observed volatilities for guideline public companies, and then warns that for early-stage companies, public peers will often be larger, more profitable, and more diversified. In that setting, the appropriate volatility may be better represented by the higher end of the guideline range over shorter time frames, moving toward the median of small public companies over longer periods. That’s the opposite of a rote downward adjustment.
The draft also supports explicit size and leverage analysis. It presents a size-decile approach showing materially different median volatilities across company sizes and notes that leverage can significantly increase equity volatility. Where peer leverage differs from the subject, normalization to asset volatility and re-levering to the subject capital structure is a more defensible move than an unexplained adjustment.
Defensible vs. red flag
The most useful frame for any valuation conclusion involving volatility is the one a reviewer is going to apply: side-by-side comparison of what looks like judgment and what looks like a lever. The same assumption can fall on either side depending on what surrounds it.
If the report can’t explain why this company, this security, and this date justify the number, reviewers see a lever where the analyst sees judgment.
Where the SEC is currently looking
Recent EDGAR materials show the staff is still working this territory. In July 2025, the SEC asked NIQ Global Intelligence plc to provide a summary of share-based awards since January 1, 2024, disclose the grant-date fair values, and reconcile those values to the midpoint of the offering range, including changes in valuation methodologies and assumptions. The company’s response disclosed a Monte Carlo simulation framework with expected volatility of 30.0% to 31.7% and changing DLOMs. Even where the comment didn’t single out “haircuts,” the staff was clearly tying pre-IPO awards to the underlying valuation mechanics.
Recent IPO filings show how live the issue is. Figma’s 2025 S-1 disclosed that, because the company lacked its own trading history, expected volatility was estimated from comparable public companies. A September 2025 confidential draft S-1 for BDF Holding Corp. similarly said expected stock volatility was based on an average of peer public-company volatilities until the issuer had adequate trading history of its own. Peer-based volatility selection is a current, disclosed judgment area in active filings.
A five-step process that holds up
A defensible process usually looks better than a defensive explanation written after the fact. The work sequence the AICPA draft and SAB 107 both point toward, in operating order:
| Step | What it requires | |
|---|---|---|
| 01 | Build the peer set | Industry, stage, size, business model, leverage. Inclusion and exclusion logic stated for each company. Index selection used for identification, not as input. |
| 02 | Align horizon | Expected term for Black-Scholes-style; contractual term where appropriate; scenario horizon for Monte Carlo. Implied volatility weighted by remaining-life comparability. |
| 03 | Address leverage | Where peer leverage differs materially, normalize equity volatility to asset volatility and re-lever to the subject capital structure. State the adjustment explicitly. |
| 04 | Reconcile to transactions | Recent financings, secondary sales, capital-stack indications. Build the bridge from observed evidence to model output. Don’t leave the reviewer to infer it. |
| 05 | Document in reviewer language | Auditors, boards, and counsel ask the same five questions. Answer them in the report before they have to be raised. |
The five questions a defensible report answers up front
- Why are these the right peers?
- Why is the selected point within the range more representative than the median or mean?
- What changed from the prior valuation date?
- How does the selected volatility square with recent transactions or calibration evidence?
- If grants occurred while the company held positive MNPI, did SAB 120 affect the current-price or volatility analysis?
The takeaway
There is rarely one right volatility number, but there is a clear difference between supported judgment and unsupported value management, and it shows up in the surrounding work rather than the input itself. Rather than blessing a rote haircut, SAB 107 requires an estimate that reflects what marketplace participants would use, supported by sufficient information and applied consistently. The AICPA draft supports fact-specific adjustments for stage, size, leverage, and horizon, and explicitly warns that early-stage companies often need higher volatility, not lower. SAB 120 says the moments most likely to attract scrutiny call for tighter discipline, not looser assumptions. A report that answers the five reviewer questions before they’re asked usually travels much better through audit, board, and transaction review than one that has to be defended after.