Four people incorporate a company on a Monday. They split the common stock evenly, four ways, each subject to a four-year vest with a one-year cliff. On Wednesday they hire their first engineer and grant her two percent of restricted common, same vesting terms. On Friday of that week their seed round closes. $40M at a $200M post-money, led by a tier-one fund that moved in 12 days from first meeting to wire.
Three of the four founders file an 83(b) election within 30 days of the Monday grant. One forgets. The engineer’s lawyer is on vacation and her paperwork goes in on day 32.
The two who filed on time will pay tax on their founder stock at the value it had on Monday, essentially zero. The two who missed the window will pay ordinary income tax at every vesting tranche, at whatever the stock is worth on each of those future dates, for four years, on equity they cannot sell.
The difference between those two outcomes, for each person, is a seven-figure tax bill on phantom income. [S: this is the version we walk new founders through on day one.]
What 83(b) actually does
Section 83(b) is a one-page election that changes the timing, not the existence, of the tax on restricted stock.
The default rule under section 83(a) is that when restricted stock vests, the employee recognizes ordinary income equal to the fair market value of the stock on the vesting date, minus what they paid for it. If you vest over four years on a monthly schedule, you have 48 separate tax events, each valued at whatever the stock was worth that month. As the company grows, each tranche becomes more expensive to vest into, and the tax is payable in cash even though the stock is not sellable.
The 83(b) election lets the recipient pay tax once, up front, on the full grant, at the current fair market value, and start the capital gains holding period on day one. If the current value is trivially small (founder stock at formation, pre-revenue common at a low 409A), the tax bill at election is trivially small too. Everything that happens later is long-term capital gains on sale.
The election must be filed with the IRS within 30 days of the transfer of the stock. Not 30 days from when you sign the offer. Not 30 days from when your lawyer finishes the stock purchase agreement. Thirty days from the transfer. The deadline is jurisdictional. The IRS has no authority to accept a late election, no matter how sympathetic the facts. There is no cure, no extension, no appeal. [M: jurisdictional means jurisdictional.]
This is the most consequential 30-day deadline in American tax law that most people have never heard of.
What changed
The 83(b) election has been in the code since 1969. The mechanics have not changed. What has changed is the shape of the rounds it sits next to.
Ten years ago, a typical seed round was $2M to $4M at a $6M to $10M post. Founders formed the company, tinkered for nine to eighteen months, hired their first couple of engineers on promises and ramen, and then raised. The stock at formation was worth effectively nothing. The stock when the first employees came on was still worth effectively nothing. The 83(b) election was important, but even if someone botched the timing, the numbers involved were small enough that the damage was recoverable.
That is not the market anymore. AI-era seed rounds routinely close at $30M to $100M in primary capital, at post-moneys between $200M and $1B, on companies that are eight weeks old. Pre-seed rounds that used to be friends-and-family checks are now $20M priced rounds led by multi-stage funds. Formation to first institutional check is measured in weeks, not years.1
The 83(b) election was designed around a world where the gap between grant-date value and vesting-date value accumulated slowly. That gap now opens in days. A founder who grants themselves stock on formation at a strike of effectively zero, and closes a round at a $200M post three weeks later, has approximately nine days of 83(b) window remaining. By the time that window closes, the stock in their hands is worth tens of millions on paper. If they miss the filing, every future vesting tranche is ordinary income at whatever the company is worth that month.
The first employees are in a structurally worse version of the same problem. They do not get their stock at formation. They get it when they are hired, which is increasingly after the round has closed. Their “current value” on the grant date is the post-round 409A, which is no longer trivial. An 83(b) election is still the right move for them in most cases, but the election itself now has a real tax cost, which the employee has to pay in cash, against illiquid stock, in the same year they took a pay cut to join a startup.
The old intuition, “83(b) is free, just file it,” is wrong now. The filing is still almost always correct. It is not free, and the planning around it has to be deliberate.
What this means for you
Three things almost every founding team gets wrong.
The grant date is not the signing date. The 30-day clock runs from the actual transfer of the stock, which is a function of board approval, a fully-executed stock purchase agreement, payment of the purchase price (even if nominal), and delivery of the stock certificate or book-entry record. Founders regularly believe they are “granted” on the date the term sheet is signed or the date the incorporation is filed. They are not. Then they miss the window because they were counting from the wrong day, or because the paperwork was not actually completed until two weeks later and nobody told them the clock had just started. This is the single most common 83(b) failure we see, and it is entirely preventable. [S: counting errors, every time.]
Restricted stock for early employees is not the same decision it was in 2018. When the post-round 409A is meaningful, and it is, almost everywhere, in 2026, the employee’s 83(b) election crystallizes a real tax liability in the year of grant. That can be the right call. It can also be the wrong call, depending on the employee’s other income, the company’s trajectory, the likelihood of an exit, and whether the company is willing to extend an early-exercise option instead. Handing a restricted stock grant plus an 83(b) form to a new hire and saying “just sign this, it’s standard” is not a standard of care that survives contact with a competent tax advisor.
Options are not always the answer, but they are often the better answer now. A properly structured option grant with an early-exercise feature lets the employee decide, within the 90 days after their hire, whether to exercise immediately (and file an 83(b) on the exercised shares, starting the capital gains clock at a known low number) or defer exercise until there is liquidity. That flexibility is worth a lot. A flat restricted-stock grant hard-codes a decision that should be the employee’s to make. For the first five to ten hires at a venture-backed startup, early-exercisable options are almost always the right instrument.
What to do
Four things, in order.
- Calendar the 30-day deadline from the actual transfer date, not the signing date, and calendar it in three places. One in your lawyer’s system, one in your own, one in the employee’s. If you do not know the exact transfer date, you do not have a deadline to calendar. Fix that first.2
- Have the 83(b) election drafted, signed, and mailed before you hand the employee their grant paperwork, not after. The form is one page. There is no excuse for it to be the thing that slips. Send it certified mail, return receipt requested, and keep the green card.
- If you are raising a round within 90 days of formation, have your equity plan designed around that fact. Founder restricted stock with 83(b) on day one. Early employee grants structured as early-exercisable options, not restricted stock, unless there is a specific reason to do otherwise. Have the 409A in place before the first option grant goes out. Talk to counsel before the term sheet is signed, not after.
- If you already missed a window, do not send the form anyway. A late 83(b) filing does nothing, and it creates an evidentiary record that cuts against you. Talk to counsel about whether a forfeiture-and-regrant or a cancellation-and-reissuance is feasible, what the section 409A implications are, and whether the company’s cap table can absorb the cleanup. Some of these situations are fixable. None of them are fixable by pretending.
If you’re about to make your first hire and haven’t structured the equity grant yet, the next thirty days matter more than most founders realize.
Talk to a Talairis attorney →A closing thought
Your first hire has 30 days.
So does everyone else at the table, including you. The window was always 30 days. What changed is what sits on the other side of it. A round that closes before the paperwork does. A 409A that is no longer trivial. A tax bill that is no longer theoretical. The companies that get this right will do it because someone, founder or GC or outside counsel, understood that the calendar is the product here, and treated it that way.
Everyone else will find out on the second anniversary of the grant, when the first large vesting tranche hits and the W-2 arrives with a number on it they cannot pay.
- Some startups sign term sheets before they have a bank account. The compression of the formation-to-funded calendar in the AI era has flipped the old “83(b) is free” intuition. The election is still right. The planning around it is no longer optional. ↩
- Every restricted stock grant carries a 30-day clock, and the clock runs from a date that almost nobody calendars correctly. ↩
