409A valuation explained: how startups set option strike prices

Kestrel, our made-up software company, has just closed a seed round and wants to hire its first engineer with an offer of stock options. The offer needs one number before it can be written: the price at which the engineer may buy shares. That number is not a negotiating position, and it is not a share of the last round’s price. It comes from a valuation that has a name, a rulebook and a shelf life. This page follows Kestrel through it. It belongs to the equity and finance team section, it describes United States federal rules, and it is education, not tax advice.

What a 409A valuation is and how it works

Public companies are valued by the price their stock trades at, but a private company needs a valuation to find the fair market value of its equity⁵. The 409A valuation does that job for the common stock, and it also fixes the strike price, which has to be at or above that value⁵. The name comes from section 409A of the Internal Revenue Code, which the American Jobs Creation Act of 2004 added, as IRS Notice 2005-1 records³. The statute’s own title is “Inclusion in gross income of deferred compensation under nonqualified deferred compensation plans”².

The rule in plain words: an option to buy company stock is not deferred compensation if the exercise price can never be less than the stock’s fair market value on the day the option is granted, and the number of shares is fixed on that day¹. The option must also be taxed under section 83, and it may carry no deferral feature beyond postponing income until exercise or disposition¹. Pass those tests and the option stays outside 409A. Miss the first one and an option with an exercise price below fair market value generally counts as deferred compensation^{1, 3}.

The test targets nonqualified options: an incentive stock option under section 422 and an option under an employee stock purchase plan under section 423 are carved out¹. The valuation exists to establish safe harbor status with the IRS⁶.

For Kestrel the question is therefore narrow. What is a share of its common stock worth on the day the engineer’s option is granted, and can the company prove it?

Why the strike price cannot be below fair market value

The reason is what happens to the holder. If a plan fails 409A, all compensation deferred under it for the year and all earlier years is includible in gross income for that year, to the extent it is not subject to a substantial risk of forfeiture and not previously included². The tax for that year also goes up by an amount equal to 20 percent of the compensation included in gross income². Interest follows, at the underpayment rate plus 1 percentage point on the underpayments that would have occurred had the amount been included when first deferred².

Run the numbers on a small grant. Say 10,000 options vest in a year, the grant set the exercise price at $0.90, and fair market value at the end of that year is $4.00. Those inputs are made up. The table applies the statute’s two amounts to them and ignores the interest, which depends on dates and rates.

Worked example (our arithmetic): a discounted option that fails the test (inclusion and 20 percent rule from²)

That is the whole case for a defensible valuation: the price protects the person who holds the option.

How is valuation calculated for a startup? Three steps

A startup’s 409A valuation is calculated in three steps⁴. The first step finds how much the company is worth, the enterprise value⁴. The second step allocates that value across the equity classes to reach the fair market value of the common stock⁴. The last step applies a discount, because the stock is not publicly traded⁴.

The Treasury rule does not name these steps. For stock not readily tradable on a securities market, fair market value is the value found by the reasonable application of a reasonable valuation method¹. Whether a method is reasonable depends on the facts and circumstances as of the valuation date¹. The factors include the value of the company’s tangible and intangible assets, the present value of its anticipated cash flows and the market value of similar corporations¹. Recent arm’s length transactions in the stock, control premiums and discounts for lack of marketability also count¹. There is no universal formula for the value of an illiquid, non-controlling interest in a closely held company⁵.

A company may use a different valuation method for each separate action¹. It must use a single method for each action and cannot alter it retroactively once used¹.

Two limits stop a method from being reasonable. It must take into account all available information material to the value of the company¹. And a value calculated earlier is not reasonable later if it misses information that came after, such as the resolution of material litigation or the issuance of a patent¹.

Step one: the value of the company

Valuation work uses three main approaches, and they can be combined: market, income and asset-based⁴. Early-stage companies tend to lean on a market approach, and more mature ones on an income approach⁴.

In the market approach, profitable companies are usually valued on an EBITDA multiple, and early-stage companies with negative EBITDA on revenue multiples⁴. Comparable sets of five or more listed companies are typical inputs⁴. The asset approach also gets expensive, because intangible assets such as IP are costly to appraise⁴.

The backsolve is what Kestrel will meet first. After a fundraise, determining the enterprise value is straightforward, because the value of that round is used⁴. A backsolve is the way a founder will almost always first meet a 409A valuation⁴. It rests on the equity value implied by the sale price of shares in a recent financing, run through an option pricing model that considers the capital structure and the rights of the preferred and common stockholders⁵.

A practical note. The forecast the valuer receives should be the one the board sees, because that forecast is one of the first places a reviewer looks to test the assumptions⁴.

Step two: from company value to common stock value

Only a company with nothing but common stock gets a one-line answer: the enterprise value divided by the fully diluted shares outstanding, and that case is extremely rare for a venture-backed company⁴. Kestrel has preferred stock held by its seed investors, so the value must be split.

OPM is the most frequently used of the four in venture-backed work, and Black-Scholes is the most commonly used option-pricing model inside it⁴. A handful of assumptions drive its output: the enterprise value, volatility and the expected time to exit⁴. The longer the time to exit, the more valuable the option, because more time gives it a better chance of finishing in the money⁴.

The allocation takes into account each class’s economic rights, meaning liquidation preference, conversion, dividends and participation⁴. The preference stack shows up in the result. A company acquired for less than or equal to its liquidation preferences leaves $0 for the common stockholders in the OPM⁴. Common stock values tend to be higher under PWERM, because in an IPO the preferred stock converts to common and the liquidation preferences disappear⁴.

Why are 409A valuations so low?

Because a common share is a different security from the preferred share the investors bought. The preferred stands in front of the common in a sale, and the allocation step prices that gap⁴. Valuations that sound too low are often reasonable because of the economic rights of the preference stack⁴. Common stock is also illiquid, which the last step prices separately⁴.

Step three: the discount for lack of marketability

The last step applies a discount for lack of marketability, DLOM, to the common stock value from step two⁴. The allocation models assume stock that can be sold immediately, and employees cannot sell theirs⁴. The main variable is the time to a liquidity event⁴. Most DLOM studies reach a discount of 25%-35% for a two-year holding period⁴. The discount shrinks as the company scales and gets more likely to succeed⁴. After a backsolve it tends to be lower than those ranges, because the valuation benchmarks to a preferred instrument that already carries some lack of marketability⁴.

Suppose step two leaves the common stock at $1.00 a share and the valuer picks a 30% discount, inside the published range. Both numbers are assumed.

Worked example (our arithmetic): the discount step only (DLOM range from⁴)

Real work is messier, because step two never gives a round number. The point is the order: allocation first, discount last.

Why hire a 409A valuation firm

A company can try to value its own stock, but the easiest and most common way to get the safe harbor is a qualified, independent valuation provider⁴. The Treasury rule gives that route a presumption. The appraisal must be independent and meet the requirements of section 401(a)(28)(C)¹. If it is made no more than 12 months before the transaction it supports, the value is presumed reasonable¹. The Commissioner can rebut the presumption only by showing that the method, or its application, was grossly unreasonable¹.

A second route exists for young companies. A valuation made reasonably and in good faith, evidenced by a written report that takes the regulatory factors into account, also qualifies for illiquid stock of a start-up¹. A start-up here means a company with no material trade or business conducted for 10 years or more and no class of equity securities traded on an established securities market¹. The route is closed when the company can reasonably expect a change in control within 90 days, or a public offering within 180 days, after the action the valuation supports¹.

The valuer must also be qualified. A valuation is not made in good faith unless the person doing it is one the company reasonably determines is qualified¹. Significant experience generally means at least five years in business valuation or appraisal, financial accounting, investment banking, private equity, secured lending or comparable work¹.

The provider does the heavy lifting, matches the analysis to the company’s narrative, helps with messaging to employees and builds a coherent valuation history⁵. The conclusion is a recommended fair market value per share of common stock, used to support option strike prices⁵. Hiring a firm still does not make the result bulletproof. An expert does not automatically guarantee the safe harbor⁴. How to choose: look for a provider with experience valuing companies like yours⁴.

What you give the valuer

The inputs are ordinary documents and estimates:

• The current charter, also known as the certificate of incorporation⁴.
• The most recent cap table⁴.
• The board presentation and recent pitch deck, if a fundraise just closed⁴.
• Company historicals plus a three-year projection of profit and loss, cash and debt⁴.
• An estimate of the options the company expects to issue over the next 12 months⁴.
• A list of five or more public comparables, which carries into later valuations⁴.
• Timing expectations for liquidity events such as an IPO or acquisition⁴.
• Significant events since the last valuation⁴.

How long it takes

With every input in hand, a final draft for board approval takes about two weeks⁴. A typical timeline has data collection, modeling, draft schedules and management review in the first two weeks, then board approval and option grants in the third week⁴. The auditor should join the kick-off call before a new valuation starts⁴. An auditor review can add time: if the valuation is complex, or an IPO is expected within two years, the auditor may ask a valuation specialist to review a draft, which adds a minimum of one week⁴. Once annual revenue reaches around $10M, investors expect audited financials, and the auditor will review the valuation because it drives stock compensation expense⁴. If the auditor finds a flaw, the company must revisit that valuation and the prior ones⁴.

How often to do a 409A valuation

The rule gives a hard stop for old numbers. A previously calculated value is not reasonable later if it was calculated for a date more than 12 months before the date it is used¹. It is also stale sooner if material new information has arrived¹. Practitioners summarize it as at least once every 12 months, or when a material event occurs, whichever comes first⁴.

Material events fall into a few groups.

A company credibly approaching an IPO values far more often, for example quarterly or even monthly⁴.

A term sheet resets the clock

The sharpest event is a signed term sheet for a priced round. Most valuation firms then treat the current 409A price as unusable for option grants, because they view the term sheet as a material event that implies a change in the company’s valuation⁷. The practical advice is to clear any promised-option backlog before signing⁷. Promised options are options, restricted stock awards or purchases, RSUs, SARs, warrants and similar securities that were promised but not yet granted⁷. A company that handles them this way should have none left to worry about⁷.

Kestrel’s own timeline shows how it plays out. If the latest valuation set fair market value at $0.05, new hires get a strike price of $0.05⁶. Suppose its valuation was done on January 1 and it then closes a pre-seed round at $8 million pre-money in July: the January number is out of date for the new hires, and a new valuation is needed⁶. In NASPP’s example the fair market value rose from $0.05 to $0.25 over that round⁶.

How much a 409A valuation can cost

For many early-stage private companies, third-party valuations often fall in the low thousands of dollars⁵. Costs can rise to multiples of that for later-stage companies, or where complexity and frequency go up, for example where the 12-month safe harbor no longer applies because an IPO is planned⁵.

What moves the price⁵:

• Stage. Later-stage businesses with more stakeholders require more work.
• Capital structure. Multiple preferred classes, participating preferred, warrants, SAFEs, convertible notes, secondary transactions and complex option plans increase effort.
• Rounds and terms. More rounds and bespoke terms mean deeper modeling.
• Rush requests. Expedited turnarounds frequently carry extra fees.
• Audit support. Workpapers beyond the report widen the scope.

The cost is small next to what it protects, which is the employee’s tax position under the rule above.

Myths about the strike price

Safe harbor is proof that the company calculated a reasonable fair market value of its common stock⁵. A valuation is not an exercise in maximizing the minimum, where the goal is for the common stock to appreciate as little as possible from the prior value⁴. Valuations are also connected to each other, so the current one should not be read in a vacuum⁴.

Myth: common stock is worth 20% of the last round’s price

A rule of thumb used to be common: set the common value at 10-20% of the latest preferred round⁴. Only in rare instances is that legitimately right, even for seed-stage companies⁴. Without running a real valuation no one can know the common stock’s fair market value from memory⁴. For Kestrel, the seed round’s price is an input to the backsolve, not a multiplier.

Myth: you can show the valuer a lower forecast

The forecast given to the valuer must be the one the board uses, and a mismatch causes problems with auditors and can put the safe harbor at risk⁴.

Myth: the lowest strike price wins

Companies want a low common value because it makes options look more lucrative⁴. If a high value starts to hurt recruiting, a company can consider a stock split to bring the strike price back down⁴. The gain from squeezing the price is small. In a six-year example with 100,000 options and a $50 share price at exit, a $0.35 strike price yields $4,965,000 before taxes⁴. A $2.00 strike price yields $4,800,000⁴. The strike prices differ by 471%⁴. The outcome differs by 3%⁴.

Myth: the strike price must equal the valuation exactly

The strike price cannot be lower than the valuation’s fair market value, but it can be higher⁴. Setting it higher is useful when the company’s enterprise value has declined since the last valuation⁴. The company can then keep the strike price flat at the earlier valuation to protect morale⁴.

Changing an option after the grant

Repricing is its own trap. A modification of the terms of a stock right is considered the grant of a new stock right¹. A modification includes any change that may give the holder a direct or indirect reduction in the exercise price¹.

When a low value backfires

A too-aggressive valuation has a shelf life. When an audit or a financing forces a clean valuation, the fair market value steps up sharply⁴. Employees hired days apart then hold very different strike prices, and resentment can build in a world of “haves” and “have-nots”⁴. In an acquisition, the buyer reviews the valuations in diligence, and a buyer who distrusts them can require the company to indemnify it for the risk⁴. Before an IPO, the SEC reviews option grants for the 12-18 months before the offering⁴. A wide gap between the strike price and the IPO price lets the SEC decide the options were granted below fair market value⁴. The valuation also feeds stock compensation expense, which is why auditors read it⁴.

The rules at a glance

Where these rules meet the tax treatment of the option itself, the statutory detail is in startup stock options. How SAFEs and notes turn into the preferred stock a valuation has to price is covered in SAFE notes, and how the round that anchors the backsolve works in how venture rounds work.