Pre-OBBBA QSBS was binary. You held for five years, or you got nothing.

That's over.

The One Big Beautiful Bill Act turned the five-year cliff into a three-step curve. For qualified small business stock issued on or after July 5, 2025, the QSBS holding period now unlocks a 50% exclusion at three years, a 75% exclusion at four years, and the familiar 100% exclusion at five. The five-year cliff is still there for anyone holding older stock, but for founders acquiring QSBS today, the math on when to exit is fundamentally different.

Most founders and advisors haven't caught up. And the ones who have are still making a subtle mistake on the tax rate — one that under-states federal tax by roughly a third. This post walks through what the new qsbs holding period tiers actually mean, the 28% rate that applies to the non-excluded half of a partial exclusion, and the decision framework for whether to sell at year 3, wait to year 4, or hold to five.

The QSBS holding period: old cliff vs. new curve

Under legacy §1202, QSBS was a compliance binary. Stock held five years or less got zero exclusion on gain. Stock held more than five years got up to 100% exclusion, capped at $10M per taxpayer or 10× basis — whichever was greater. There was no middle ground and no partial credit.

OBBBA restructured that. For stock acquired on or after July 5, 2025:

  • Less than 3 years held: 0% federal exclusion.
  • 3 years to less than 4 years held: 50% federal exclusion. Per-taxpayer cap is the greater of $15M or 10× basis.
  • 4 years to less than 5 years held: 75% federal exclusion. Per-taxpayer cap is the greater of $15M or 10× basis.
  • 5+ years held: 100% federal exclusion. Per-taxpayer cap is the greater of $15M or 10× basis.

Stock acquired on or before July 4, 2025 keeps the legacy rules — 5-year cliff, $10M cap, $50M gross-asset threshold at issuance. Post-OBBBA stock gets the new tiers, the higher $15M cap, and a $75M gross-asset threshold.

Which means one cap table can have two QSBS regimes sitting side by side. A founder who took shares at formation in 2023 and then received additional QSBS in a 2026 refresh grant has two different vintages to track, each with its own clock and exclusion rules. Founders and CFOs should map this at the grant level before a liquidity event, not during one.

A quick note on what counts as "acquired." The QSBS clock starts when you actually own the stock. For founders with early-exercise rights, that's the exercise date (with a properly filed 83(b) election). For employees, it's the exercise of options, not the grant date. Options sitting unexercised don't accrue QSBS time.

The 28% rate that changes the math

Here's the part almost nobody gets right in founder-facing content.

When you take a partial exclusion under the 3-year or 4-year tier, the non-excluded portion of your gain isn't taxed at the normal 20% long-term capital gains rate. It's taxed at 28%.

The statutory path: IRC §1(h)(7) defines "Section 1202 gain" as the portion of gain that would have been excluded under §1202 but for the percentage limitation. §1(h)(4)(A)(ii) rolls that Section 1202 gain into the "28-percent rate gain" bucket. §1(h)(1)(F) caps the rate on that bucket at 28%.

Then the 3.8% Net Investment Income Tax (§1411) applies on top. Combined federal rate: approximately 31.8% on the non-excluded portion of a partial QSBS exclusion.

Holland & Knight put it directly in their July 2025 OBBBA analysis:

"The new tiered exclusion amounts also bring back into play the 28 percent capital gains tax rate for the portion of the maximum gain taken into account under Code Section 1202 that is not actually excluded."
— Holland & Knight

Greenberg Traurig is equally explicit: "Gain subject to the partial 50% and 75% exclusions are subject to tax at a 28% federal income tax rate and the net investment income tax of 3.8%."

Most founder-facing articles miss this. Many say something like "the non-excluded portion is taxed at your regular long-term capital gains rate" — ~23.8% with NIIT. That's wrong, and it under-states the tax bill by about $800K on a $10M non-excluded gain.

Blended federal rates on total gain:

  • 3-year tier: 50% excluded, 50% taxed at ~31.8% → ~15.9% blended federal rate.
  • 4-year tier: 75% excluded, 25% taxed at ~31.8% → ~7.95% blended federal rate.
  • 5-year tier: 100% excluded, 0% taxed → 0% blended federal rate.

One good piece of news: post-OBBBA 3-year and 4-year exclusions do not generate the AMT preference item — 7% on the legacy 50% tier and 42% on the legacy 75% tier — that plagued the pre-September 28, 2010 partial exclusions. OBBBA cleaned that up. The AMT drag is gone.

When year 3 is the right exit

A 15.9% blended federal rate on what would otherwise be ordinary QSBS gain is still attractive — especially against the baseline of "no exclusion at all" that legacy QSBS imposed on pre-5-year exits. The 3-year tier is the right call in several situations.

Forced exits. An acquirer shows up with a term sheet 38 months after issuance. The board signs a definitive agreement at month 39. Closing happens at month 42. You're not choosing to exit early — the exit chose you. Taking the 50% exclusion plus paying ~15.9% blended federal on the rest is dramatically better than the pre-OBBBA alternative, which was zero exclusion and full ~23.8% LTCG on the whole gain.

Concentration risk. You founded a company at 28 and you're turning 31 with more than 90% of your net worth in one illiquid stock. A tender offer or secondary at year 3 gives you the chance to rebalance. The ~15.9% blended cost is the price of diversification.

Life reasons. House, divorce, family health event. Founders sometimes need the money. A year-3 exit with real liquidity is better than a year-5 exit that doesn't happen because circumstances changed.

Worked example — year 3

You acquired QSBS in September 2025 at a basis of $50,000. Three years later, an acquirer offers you $20M in cash for the shares.

  • Gain: $20M − $50,000 ≈ $19.95M
  • 3-year tier: 50% excluded = ~$9.98M tax-free
  • Remaining ~$9.98M taxed as §1202 gain at ~31.8% = ~$3.17M federal tax
  • Net after federal tax: ~$16.83M

If you're in a non-conforming state like California, layer in another $20M × 13.3% = ~$2.66M of state tax. Now the year-3 net drops to ~$14.17M. More on state tax below.

When waiting to year 4 or year 5 wins

The 3-year tier is a real benefit. But every year you hold past 3 compounds into a meaningfully lower tax bill.

From year 3 to year 5 on a $20M gain, the spread is ~$3.17M in saved federal tax — roughly 16% of total gain. That's substantial money in exchange for two years of additional illiquidity and concentration.

Year 4 is the interesting compromise point. On the same $20M gain:

  • 4-year tier: 75% excluded = $15M tax-free (or up to the cap)
  • Remaining $5M at ~31.8% = ~$1.59M federal tax
  • Net after federal tax: ~$18.41M

The year-3 → year-4 improvement is ~$1.59M. The year-4 → year-5 improvement is another ~$1.59M. The curve is roughly linear — each additional year of holding buys you a similar chunk of net gain.

Hold to year 5 when:

  • You control the exit timing (planned IPO window, scheduled tender, no acquirer pressure)
  • Your personal concentration risk is manageable — other liquid assets, low personal burn
  • You have strong conviction in the company's continued appreciation
  • You're within months of the 5-year mark — don't trigger year 3 or 4 if year 5 is close

Consider year 4 when:

  • Year 5 isn't reachable (acquirer won't wait, you need liquidity)
  • But year 4 is close enough that waiting a few extra months is doable
  • A 4-year partial exit + §1045 rollover on the non-excluded portion can preserve some of the year-5 benefit on replacement QSBS

Year 3 vs. year 4 vs. year 5, side by side (on a $20M gain)

  • Year 3 exit: ~$3.17M federal tax → ~$16.83M net. Spread vs. year 5: $3.17M.
  • Year 4 exit: ~$1.59M federal tax → ~$18.41M net. Spread vs. year 5: $1.59M.
  • Year 5 exit: $0 federal tax → ~$20.00M net. Spread vs. year 5: $0.

In a non-conforming state, you still owe state tax on 100% of the gain no matter which year you exit — the state-tax line is flat across all three years. Which makes the state-tax angle its own decision input.

The decision framework

When you're staring at an actual liquidity event at year 3 or 4, the right question isn't "can I take a partial exclusion?" It's "what does my specific situation say about selling now versus waiting?"

The variables that matter:

  • Exit certainty. Is the acquirer committed? Signed LOI? Closed funding? Or still speculative?
  • Replacement target. If you roll §1045, do you have a credible replacement QSBS identified? Or would you be parking proceeds in a shell corporation (which, per John P. Owen v. Commissioner, the Tax Court doesn't bless)?
  • State tax exposure. Living in California, Pennsylvania, Alabama, Mississippi, or Hawaii? State tax hits the full gain regardless of your federal tier.
  • Personal concentration. What percentage of net worth is in this one illiquid stock?
  • Alternative uses of proceeds. Is there a compelling reinvestment, debt paydown, or trust-funding opportunity that argues for earlier liquidity?
  • Life factors. Family, health, timeline for a next venture.

A rough decision framework, by situation:

  • Forced exit, no control, post-OBBBA stock. Take the partial exclusion at whatever tier you hit. Consider a §1045 rollover on the non-excluded portion if you have a replacement target.
  • Voluntary sale, flexible timing, year 4 reachable. Wait to year 4 if you can. The year-3 → year-4 jump is ~$1.59M on a $20M gain.
  • Voluntary sale, year 5 reachable, no liquidity pressure. Wait to year 5. The 100% exclusion is the whole point of QSBS.
  • Heavy concentration, year 2.5, tender offer available. Take the offer if you can reach year 3 by close. Partial beats zero.
  • Non-conforming state, any tier. Layer state tax into the decision. The federal tier math alone under-states your real cost.

The §1045 rollover is the escape hatch that ties this all together. If you're forced out at year 3 but you have a replacement QSBS you believe in, you can roll the non-excluded portion under §1045, defer the ~$3.17M of federal tax, and continue the holding clock toward a full §1202 exclusion on the replacement stock. It's the reason "sell at year 3" doesn't have to mean "lock in the 15.9% rate."

What founders get wrong

Even founders who know the tier structure routinely make one or more of these mistakes.

Assuming the regular LTCG rate on the taxable half. Back-of-envelope math at ~23.8% on the non-excluded portion under-states federal tax by roughly a third. The correct rate is ~31.8% on that portion — so the effective blended rate at year 3 is ~15.9%, not ~11.9%. This is the single most common error, and it changes the year-3 vs year-5 decision meaningfully.

Forgetting state tax. California doesn't conform to §1202 and hasn't since 2013. Pennsylvania, Alabama, Mississippi, and Hawaii also don't fully conform. If you live in one of these states, the state taxes 100% of your gain at the state's rate regardless of your federal tier. A California founder with a $20M gain pays ~$2.66M in state tax whether they exit at year 3, 4, or 5. That's a flat cost on top of the federal curve — and it often makes concentrated-founder relocations worth considering well before exit. See our state-tax implications guide for the full breakdown.

Treating OBBBA tiers as automatic. The 3/4/5-year tiers only apply to stock acquired on or after July 5, 2025. Stock you already hold from earlier grants is still under the legacy 5-year cliff. Cap tables with mixed vintages are common; the math is different for each tranche.

Missing the 83(b) timing. If you exercised options or received restricted stock but didn't file an 83(b) within 30 days, your QSBS clock for unvested shares starts as shares vest, not when you signed. That can push your "year 3" milestone out by years. See our QSBS primer for more on what actually starts the clock.

Conclusion

OBBBA turned QSBS into a curve, not a cliff. But the curve only works if you do the math on the real rate.

Three things to take away:

  • The non-excluded portion of a partial §1202 exclusion is taxed at ~31.8%, not ~23.8%. Blended federal rates are ~15.9% at year 3, ~7.95% at year 4, and 0% at year 5.
  • The year-3 → year-5 spread on a $20M gain is roughly $3.17M. Whether that's worth two years of illiquidity depends on your exit certainty, concentration, and state-tax exposure.
  • State tax doesn't tier. Non-conforming states tax the full gain at whatever year you sell, so the real decision is federal-curve-plus-flat-state-tax.

If you're evaluating an exit in the next 12–24 months and want to model what each scenario actually looks like — including §1045 options for early exits and trust-stacking implications on the federal side — schedule a call. We'll walk through your specific cap-table vintages, state-tax exposure, and liquidity scenarios so you can decide with real numbers, not back-of-envelope estimates.

‍