How to Increase Your QSBS Exclusion with Trusts

If you hold QSBS acquired on or before July 4, 2025 and have cleared the five-year mark, you are eligible for the exclusion, but only up to $10 million per person, per issuer (or 10 times your adjusted basis, if greater). For a founder with a $50 million exit and zero-basis stock, that leaves roughly $40 million exposed to federal tax at 23.8%, or about $9.52 million owed to the IRS on gain the law was designed to protect.

Stock acquired after July 4, 2025 carries a higher $15 million cap under the One Big Beautiful Bill Act (OBBBA), but that stock cannot reach a full 100% exclusion until July 2030 at the earliest. The higher cap is where the rules are headed; the $10 million cap is the scenario most commonly faced by founders exiting in 2026, though stock vintage governs, and earlier-acquired shares require separate analysis.

Trusts are one of the primary tools for getting more of that gain excluded. When structured correctly, each trust holds its own exclusion cap, and you can stack multiple trusts to cover a larger share of the total gain.

Scope note: Unless otherwise noted, the pre-OBBBA examples in this article assume qualifying stock acquired between September 28, 2010, and July 4, 2025, held for more than five years, with no prior use of the applicable per-issuer dollar limit. Stock acquired before September 28, 2010 may qualify for only a 50% or 75% exclusion and may generate an alternative minimum tax preference item. Earlier-vintage stock requires separate analysis.

The QSBS exclusion under Section 1202 is a per-taxpayer, per-issuer cap. The cap is the greater of the flat dollar amount or 10 times your adjusted basis in the stock. Gifted founder stock carries the donor’s near-zero basis, so typically, trust stacking with founder stock relies on the flat cap, not the 10x alternative.

The applicable flat cap depends on when the stock was acquired:

The OBBBA, signed July 4, 2025, made three changes that matter here:

One more constraint that surprises many founders: spouses filing jointly generally share one combined QSBS exclusion under Section 1202(b)(3). Adding your spouse as a holder does not add an additional exclusion. Additional QSBS exclusions come from additional separate taxpayers, which is where trusts come in.

Here is what each additional Exclusion is worth, assuming a 100% QSBS exclusion and zero basis:

Savings measured against the 23.8% federal rate (20% LTCG plus 3.8% NIIT) that would otherwise apply to non-excluded long-term gain. Dollar-limit examples assume no prior utilization for the same issuer and that the 10-times-basis alternative does not produce a larger limit.

Section 1202 treats each taxpayer as entitled to their own QSBS exclusion cap per issuer. A properly structured non-grantor trust is treated as a separate taxpayer for Section 1202 purposes. That means each trust that holds qualifying QSBS is eligible for its own exclusion.

The core mechanic works like this:

Under Section 1202(h), the gift transfers the stock’s QSBS character to the trust, and the trust tacks the donor’s holding period and acquisition date.

Not every trust works for this strategy. The trust must be structured as a non-grantor trust.

A grantor trust is transparent for tax purposes. Under Rev. Rul. 85-13, the IRS attributes the income directly to you, which means the trust does not get its own exclusion cap. It uses yours. For this strategy to work, the trust must stand on its own as a taxpayer.

Irrevocability alone is not enough. Irrevocable and non-grantor are separate concepts: a trust can be fully irrevocable and still be a grantor trust if the grantor or the grantor’s spouse retains the wrong powers or interests. The most common trap is the Spousal Lifetime Access Trust. A standard SLAT is typically a grantor trust under Section 677 because the spouse is a discretionary beneficiary, and it does not create a separate exclusion cap. A non-grantor SLAT variant (referred to as a SLANT) requires deliberate, specialized drafting.

The following are structural best practices and design features that support non-grantor status; they are not a statutory checklist. Grantor trust status is determined under Sections 671 through 679 based on the actual powers and interests involved.

What makes a trust a non-grantor trust for QSBS purposes:

Separate trustees, EINs, and administrative details support the structure but are not, by themselves, a guarantee of separate exclusions. Whether any trust is treated as a separate taxpayer depends on the full analysis of grantor status, beneficial ownership, retained powers, and potential trust aggregation under applicable law. Counsel should evaluate each trust independently.

Stacking is squarely on Treasury’s radar. In May 2026, Kenneth Kies, Treasury’s Assistant Secretary for Tax Policy and acting IRS Chief Counsel, publicly warned that guidance limiting QSBS stacking is coming (“Let me just warn you: We don’t like stacking”), echoing earlier remarks by a Treasury attorney-adviser. As of this writing, no formal guidance specifically restricting QSBS trust stacking has been issued, and any structure should be reviewed against whatever authority exists at the time of implementation.

The existing statutory hook is Section 643(f), which allows the IRS to treat two or more trusts as a single trust when they have substantially the same grantors and substantially the same primary beneficiaries, and a principal purpose is income tax avoidance. An existing multiple-trust regulation under Section 643(f) is already in force; whether it applies directly to Section 1202 exclusions is contested, and Treasury may rely instead on its regulatory authority under Section 1202(k). Either way, the practical target is clear.

Changing trustees, EINs, or administrative details does not, by itself, resolve the aggregation analysis. Conversely, using the same professional corporate trustee does not, by itself, establish that trusts must be aggregated.

The fact pattern that draws fire is the carbon copy trust: multiple trusts that are structurally identical, with the same terms, the same trustee, and the same beneficiaries, created for no reason other than to manufacture extra taxpayers and multiply the cap. Treasury officials have specifically flagged arrangements that layer overlapping combination trusts (a trust for child A, a trust for child B, then an AB trust, a BC trust, and so on) purely to add exclusions.

What makes a structure defensible is genuine differentiation, backed by real planning purpose:

Transferring QSBS to a trust after a sale is already in motion creates substantial assignment-of-income risk and may leave the gain taxable to the donor. The assignment of income doctrine governs when income is considered “earned” by a taxpayer. Transfers made too close to, or after, an agreement has been signed, and with virtually no deal contingencies left, can be treated as if the original owner earned the income. In Hoensheid v. Commissioner, T.C. Memo. 2023-34 (a charitable stock gift case applying the assignment-of-income doctrine), the Tax Court attributed gain back to donors who transferred stock two days before closing, when the sale was already practically certain.

These circumstances can create substantial assignment-of-income risk:

There is no IRS safe harbor based on a number of months or days before a sale. The test is facts and circumstances. As a practitioner heuristic, not a legal rule, transfers 12 to 24 months or more before any sale process begins are far more defensible, and earlier is better. Ideally, fund the trusts when equity value is still low. This reduces the gift tax impact of the transfer and creates a clear, defensible record that the transfer preceded any contemplation of a sale.

Timing is an extremely important variable in this strategy. Late funding can defeat the intended income-tax result of the transfer.

This article focuses on completed-gift, irrevocable non-grantor trusts. Funding a trust is a completed gift, and there is a cost to making that gift. A completed taxable gift generally uses the donor’s remaining federal gift and estate tax exemption, which is $15 million per individual in 2026 (a married couple has two individual exemptions), reduced by any prior taxable gifts.

This is another reason early transfers win: gifting when the equity value is low uses far less exemption per dollar of future excluded gain than gifting on the eve of a sale.

For founders in states that do not recognize the federal QSBS exclusion, the trust strategy has a second layer of value. California, Pennsylvania, Mississippi, Alabama, Oregon (SB 1507, retroactive to January 1, 2026), and Illinois (SB 3019, effective for tax years ending on or after December 31, 2026) do not conform to the federal QSBS exclusion. New Jersey, by contrast, now conforms for tax years beginning on or after January 1, 2026.

California deserves particular care. Under Cal. Rev. & Tax. Code Section 17742, California taxes a non-grantor trust based on the residence of its fiduciaries and its non-contingent beneficiaries, not the trust’s stated situs. A Nevada-situs trust does not escape California tax simply by being formed in Nevada. The structure requires a non-California trustee AND no California-resident non-contingent beneficiaries. California-sourced income remains taxable to the trust regardless.

California Example: A founder retains 60% of QSBS personally and gifts 40% to two properly structured Nevada non-grantor trusts (each holding 20%). On a $50 million total exit:

Note: This example assumes zero basis and California’s top marginal capital gains rate of 13.3%. The trusts avoid California tax only if the Section 17742 requirements are properly designed (non-California fiduciaries and no California-resident non-contingent beneficiaries). Distributions of accumulated income to California-resident beneficiaries in later years can trigger California throwback tax. Consult qualified tax counsel to confirm application to your specific structure.

This article addresses completed-gift non-grantor trust structures. California’s treatment of incomplete-gift non-grantor trusts (ING trusts) differs; since January 1, 2023, California generally requires ING trust income to be reported by the grantor, subject to a narrow exception.

Here is a commonly faced example of a founder exiting in 2026. Founder stock acquired before July 5, 2025 ($10 million flat cap per taxpayer), zero basis, held more than 5 years, so each holder qualifies for a 100% exclusion up to its cap. The “$40 million exit” in this example refers to the proceeds attributable to the founder’s shares across all holders, not the company’s overall transaction value.

A founder anticipating a $40 million exit retains 25% personally and gifts 25% each to three properly differentiated non-grantor trusts, well before any sale process begins:

The additional exclusions come with a real ownership tradeoff. In this example, the founder retains $10 million of sale proceeds personally, while $30 million belongs to the trusts and must be administered under their terms. The potential tax savings benefit the founder and trust beneficiaries collectively; the transferred assets are not the founder’s personal spending account and are now out of the founder’s estate.

Before deciding how much to transfer, determine how much capital you need to retain for personal liquidity, taxes at close, future investments, and other family commitments. At Keystone, we run multiple planning and stress-testing exercises with our founder clients to help determine how much (if any) to transfer.

Acting alone, the founder excludes $10 million and pays roughly $7.14 million in federal tax on the remaining $30 million ($30 million at 23.8%). With the trust structure, the entire $40 million gain is excluded from federal tax. The structure is worth approximately $7.14 million in federal tax savings, before any state-level benefit.

Forward-looking hypothetical (2031 or later): the same structure holding stock acquired after July 4, 2025, once the 5-year tier is reachable in July 2030, would carry $15 million caps per holder. Four holders could shield up to $60 million of gain (using the $15 million base amount, before inflation adjustments), and each trust would be worth up to roughly $3.57 million in federal savings at a 100% exclusion. A trust selling at the 3-year tier instead would exclude only 50%, with the non-excluded portion taxed at up to 31.8%.

Trust stacking is not a last-minute tactic. It requires coordination across legal, tax, wealth management, and estate planning professionals well before a liquidity event comes into view.

The holding period clock runs from the original acquisition date. For stock acquired through an option exercise, the holding period generally begins with the stock acquisition rather than the option grant. Early-exercised or otherwise unvested shares require separate analysis, including any Section 83(b) election. A gift of shares that already meet the five-year holding period passes the exclusion benefit to the trust; the applicable exclusion percentage depends on the stock’s original acquisition date.