A working reference
Charitable remainder trusts: how they work, how they're taxed, and when they actually pay off
Most explanations of the charitable remainder trust stop at the deduction and the tax-free sale inside the trust. Those are the easy parts. The parts that decide whether the trust was a good idea — how the distributions are taxed on the way out, what the corpus surrendered to charity actually costs, and which alternative the client would otherwise have pursued — usually go unmentioned.
This page covers all of it, including the arithmetic that runs against the instrument.
Written for advisors, planners, attorneys, CPAs, and the clients they work with. Nothing here is legal or tax advice.
The structure in one paragraph
A charitable remainder trust is an irrevocable, tax-exempt split-interest trust under Internal Revenue Code section 664. The donor transfers property to the trust. The trust pays a stream of distributions to one or more noncharitable beneficiaries — typically the donor, or the donor and spouse — for life or for a term of years. Whatever remains at the end passes to one or more qualified charities. Because the trust is exempt, it can sell appreciated property without paying capital gains tax at the time of sale, and reinvest the full proceeds. The donor takes an income tax deduction at funding for the present value of the charitable remainder.
Two things follow from that description and are routinely omitted from it. The trust does not pay tax on the sale, but the beneficiary does pay tax on the distributions, and the character of that tax is determined by what happened inside the trust. And the remainder is not a residual left over by accident — it is a mandatory transfer of at least ten percent of the funding value, and usually much more, out of the family.
Illustration · round numbers
Where the money goes, step by step
A $1,000,000 position with a $200,000 cost basis. One client puts it in a unitrust at a 6 percent payout; the other sells it, pays the tax, and reinvests. Both are followed for 25 years.
- Step 1 of 8
The trust is established
The donor transfers a $1,000,000 position with a $200,000 basis. The embedded gain is $800,000. From this moment the transfer cannot be undone and the principal cannot be recovered.
- Step 2 of 8
The trust sells and diversifies
The trustee sells the concentrated position. Because the trust is exempt, no capital gains tax is paid at the sale. The client who sells outright owes about $264,800 on that gain at a 33.1 percent combined California rate.
- Step 3 of 8
A higher dollar amount stays invested
The trust reinvests the whole $1,000,000. The seller reinvests $735,200. Basis carries over to the trust under section 1015 — what differs is the amount left working. That gap is the trust's structural head start.
- Step 4 of 8
Both portfolios grow and pay out
Each is run the same way: 7 percent growth, 6 percent drawn each year, 25 years. The head start compounds. The trust corpus reaches about $1,155,000; the taxable portfolio about $850,000.
- Step 5 of 8
Income received, and taxed
The trust pays out about $1,610,000 over the period against about $1,183,000 from the taxable portfolio — roughly $427,000 more. But section 664(b) forces the $800,000 of gain out before any principal, so about $264,800 of capital gains tax is paid along the way. The same amount deferred at step 2, moved down the timeline rather than removed from it.
- Step 6 of 8
What is left at the end
The trust's remaining $1,155,000 passes irrevocably to charity and nothing reaches the heirs. The taxable portfolio's $850,000 passes to the heirs with a section 1014 basis step-up, wiping out the accumulated gain.
- Step 7 of 8
The trade, stated plainly
About $427,000 more income, against $850,000 of forgone estate, and $1,155,000 to charity. Whether that trade is worth making depends on the client. This illustration used a $200,000 basis and average life expectancy. Both of those move.
- Step 8 of 8
Move two variables and the answer inverts
The hardest comparison is not selling and reinvesting. It is keeping the asset, drawing the same income, and still leaving an intact estate with a stepped-up basis — everything the trust delivers, with the principal kept in the family. A trust that beats that has already paid for the charitable remainder. Where a client's basis sits near 10 percent and the beneficiaries' health suggests they will outlive the IRS table, that is what the published simulations project. Neither is a lever you pull. Both are facts about the client in front of you, and the only way to know is to run the case.
Which client is in front of you is an empirical question. QuantiCRUT™ runs a specific case against all three benchmarks on 10,000 simulated paths, at both a population and an annuitant mortality basis, and reports the projected win probability and the median dollar difference for each. See how the screen works →
What section 664 actually requires
A trust that misses any of these is not a charitable remainder trust, and the consequences are not partial.
- Irrevocable. Once funded, the donor cannot reach principal or unwind the structure. This is the single most consequential feature and the one most often treated as a footnote.
- A payout between 5 and 50 percent. Measured against initial fair market value for an annuity trust, and against annually revalued fair market value for a unitrust.
- A term of lives, or a term of years not exceeding 20. The two can be combined in permitted ways, but the twenty-year ceiling on a fixed term is hard.
- A remainder worth at least 10 percent of the initial fair market value, computed at funding using the section 7520 rate and the prescribed mortality table. Section 664(d)(1)(D) for annuity trusts, 664(d)(2)(D) for unitrusts. This test is what limits how high the payout rate can go for older beneficiaries.
- For annuity trusts only, a 5 percent probability-of-exhaustion test. A CRAT that has more than a 5 percent chance of running out before the term ends fails, under the analysis of Rev. Rul. 77-374. Rev. Proc. 2016-42 provides a qualified-contingency safe harbor that avoids the test. Unitrusts, whose payment floats with the corpus, cannot exhaust in the same way and are not subject to it.
- A qualified charitable remainder beneficiary. The identity of the charity affects the donor's deduction limits.
The 10 percent test is calculated at funding and is binding. It is also the one place where the section 7520 rate genuinely constrains the plan — a point returned to below.
Unitrust or annuity trust
The choice is usually presented as fixed income versus inflation hedge. That is true and incomplete. The more consequential difference is what happens in a bad market.
| CRAT (annuity trust) | CRUT (unitrust) | |
|---|---|---|
| Payment | Fixed dollar amount, set at funding | Fixed percentage of assets, revalued each year |
| Market decline | Payment does not fall; corpus absorbs it and can exhaust | Payment falls with the corpus; the beneficiary absorbs it |
| Additional contributions | Not permitted | Permitted |
| Variants | One form | Standard, net-income (NICRUT), net-income-with-makeup (NIMCRUT), and flip |
| Exhaustion test | Applies | Does not apply |
The flip unitrust deserves a specific mention because it solves a real problem. A trust funded with illiquid property — real estate, a closely held interest — cannot reliably pay a standard unitrust amount before the asset sells. A flip CRUT pays net income only until a defined triggering event, then converts to a standard percentage payout. Sale of the contributed asset is a permitted trigger provided it is not within the donor's or trustee's discretion in a way that fails the regulations.
Unitrusts remain the dominant form, though the margin has narrowed. In 2012, unitrusts were 86.2 percent of all charitable remainder trusts. By 2022 they were roughly 78 percent, with annuity trusts growing in absolute number over the same period.
The deduction: what it is, and what it is not
The donor's income tax charitable deduction equals the present value of the charitable remainder interest, computed under Treas. Reg. section 1.664-4 for unitrusts using the prescribed mortality table and, for annuity trusts, the section 7520 rate. It is a calculation, not a projection. It is correct only for the inputs supplied and only on the valuation date, and inputs go stale: the 7520 rate resets monthly, asset values move, ages advance.
Three limits bear on what the deduction is worth in cash.
- The AGI ceiling. For long-term appreciated capital gain property contributed to a trust with a public charity remainderman, the deduction is generally limited to 30 percent of adjusted gross income, with a five-year carryforward. Cash contributions and private-foundation remaindermen carry different limits.
- The 2026 rate cap. Under the One Big Beautiful Bill Act, Pub. L. No. 119-21, the tax benefit of an itemized charitable deduction is capped at 35 percent for taxpayers in the 37 percent bracket, effective 2026. A donor in the top bracket no longer deducts at the top rate.
- Substantiation. Noncash property over $5,000 requires a qualified appraisal under section 170(f)(11). For real estate and closely held interests this is not a formality; it is where deductions are lost on audit.
A counterintuitive property of the section 7520 rate is worth knowing before anyone times a trust around it. In the unitrust remainder computation the rate serves simultaneously as the assumed growth rate of the corpus and as the discount rate applied to the beneficiary's distributions. The two effects cancel, and the remainder factor is determined by the payout rate and mortality alone. The derivation is set out in Klaus Gottlieb, Opening the Black Box: The Actuarial Derivation of the CRUT Charitable Deduction (May 17, 2026), SSRN 5924942.
The practical consequence: for a unitrust, the 7520 rate does not move the deduction. It does still determine whether a given payout-and-age combination clears the 10 percent remainder test, which is a qualification constraint rather than an economic one.
The part that gets left out
How the distributions are taxed
The trust does not pay tax when it sells the appreciated asset. The beneficiary pays tax as the gain comes out. Section 664(b) prescribes the order, and it is the least forgiving ordering rule in the Code: income comes out worst first.
Ordinary income
Current-year ordinary income first, then ordinary income accumulated in prior years and not yet distributed. Interest, non-qualified dividends, rents.
Capital gain
Current-year capital gain, then accumulated capital gain — including the entire gain the trust realized when it sold the contributed asset. Within the tier, the highest-taxed classes come out first: collectibles gain, then unrecaptured section 1250 gain, then long-term capital gain.
Other income
Tax-exempt income, current then accumulated.
Corpus
Tax-free return of principal. In practice most beneficiaries never reach this tier.
Read that ordering against the usual sales illustration. A client funds a trust with a low-basis position, the trustee sells, and the whole embedded gain lands in tier two. Every distribution thereafter carries out that gain at capital gains rates — plus the 3.8 percent net investment income tax and state tax where applicable — until the accumulated gain is exhausted. The capital gains tax was deferred and spread. It was not eliminated.
That is a real benefit. Deferral has genuine present value, and spreading a large gain across decades can keep a client out of the top rate in any single year. But it is a different benefit from the one most illustrations imply, and it is smaller.
What you can fund it with, and four traps
Cash, publicly traded securities, real estate, closely held business interests, and other property of significant value can all fund a charitable remainder trust. Four categories carry problems that general descriptions omit.
Mortgaged real estate
Debt-financed income is unrelated business taxable income under section 514. A charitable remainder trust with any UBTI in a year pays a 100 percent excise tax on that UBTI under section 664(c). Encumbered property also raises grantor trust exposure under section 677 where trust income can discharge the donor's obligation, and self-dealing questions where the donor remains personally liable. Debt should generally be cleared before funding, and the clearing itself can be a taxable event.
An asset already under contract
If the donor is legally bound to sell before contributing, the assignment-of-income doctrine attributes the gain to the donor notwithstanding the transfer. Rev. Rul. 78-197 sets the standard. Negotiations in progress are generally survivable; a binding commitment is not. This is a timing question, and getting it wrong costs the entire benefit.
Tangible personal property
Artwork and similar property are commonly listed as fundable. The deduction consequences are unattractive. Section 170(e)(1)(B)(i) reduces the deduction to basis for property put to an unrelated use, and section 170(a)(3) defers the deduction for a contribution of tangible personal property in which an intervening interest is retained. In most fact patterns the deduction the donor expected does not arrive when expected, or at all.
S corporation stock
A charitable remainder trust is not a permitted S corporation shareholder. Contributing S stock terminates the election.
Two benefits that are commonly overstated
Estate tax
Marketing material frequently says the trust assets are not part of the donor's estate. That is loose, and for some structures it is wrong. A donor who retains a lifetime unitrust interest has retained the enjoyment of the property, and the trust corpus is included in the gross estate under section 2036(a)(1). The reason there is usually no estate tax is that the charitable remainder qualifies for the estate tax charitable deduction under section 2055, and where the surviving spouse is the only other beneficiary, the spousal interest qualifies for the marital deduction under section 2056(b)(8). Inclusion offset to zero is not the same as exclusion, and the distinction stops being academic the moment a non-spouse beneficiary is named — which creates a completed gift at funding unless the donor retains the right to revoke that interest by will.
Creditor protection
The corpus is beyond the donor's reach, which is not the same as beyond a creditor's reach. The donor's retained unitrust or annuity interest is a valuable property right, and a self-settled spendthrift restriction on it is generally ineffective against the donor's creditors in most states. A creditor who cannot attach the corpus can usually attach the income stream.
What it costs to run
A charitable remainder trust is not a document that is signed and forgotten. Recurring obligations and costs include:
- Annual Form 5227, the split-interest trust information return, which is publicly disclosable under section 6104(b)
- Schedule K-1 to each noncharitable beneficiary, characterizing distributions by tier
- Annual valuation of trust assets — for a unitrust this is not optional, since it sets the payment
- Trustee fees, which vary from nothing where the donor serves, to institutional schedules where a bank or charity serves
- Investment management, where the mandate to diversify is a fiduciary duty rather than a preference
- Drafting and qualified appraisal at the front end
Trustee and management costs bear directly on whether the structure outperforms. A one percent all-in annual cost compounds against the trust over a multi-decade horizon. University planned-giving programs often charge nothing; a corporate trustee will not. The comparison is sensitive to which one is actually chosen.
The question the illustrations skip
Compared to what?
Almost every charitable remainder trust illustration in circulation makes the same comparison: sell the asset today and pay the tax, or contribute it to the trust and don't. Set out that way the trust wins by construction, because the taxable side has been handed a large immediate tax bill and the trust has not.
The comparison is only valid if immediate sale is what the client would actually have done. Frequently it is not. A client who says "I would just hold it and live off it" faces a different opportunity cost. A client who says "I would hold it and leave it to my children" faces a third, and that one carries a basis step-up at death under section 1014 that the trust cannot replicate.
Three alternatives are worth naming, because the choice among them often decides the recommendation:
- Hold and draw. Keep the asset, withdraw the same income from it, pay capital gains tax on the gain portion of each withdrawal, and pass the residual to heirs with a stepped-up basis. This is the hardest comparison for a trust to beat, because it delivers equivalent income and preserves the estate.
- Hold to death. Keep the asset untouched, take no income, and pass it intact with a stepped-up basis.
- Liquidate and reinvest. Sell now, pay the tax, reinvest what is left. This is the benchmark the standard illustration uses, and it is the one the trust most easily beats, because the alternative starts with a permanent corpus deficit.
Peer-reviewed simulation evidence puts numbers on how much the choice matters. Running 10,000 matched market paths against each benchmark for a baseline profile — a couple aged 63 and 65, a $1 million asset with a 20 percent basis, a 6 percent payout, 20 percent portfolio turnover, California rates — the trust is projected to produce more present-value personal wealth than hold and draw in 28.2 percent of paths, than hold to death in 57.8 percent, and than liquidate and reinvest in 66.4 percent.
Same client. Same trust. Three answers, spanning 38 percentage points. The benchmark is not a modeling detail; on these figures it is frequently the decisive assumption.
Klaus Gottlieb, “When Does a Charitable Remainder Unitrust Outperform? A Monte Carlo, Multi-Benchmark Suitability Framework,” Journal of Financial Planning 39 (8): 60–79 (August 2026). Open access.
All figures above are projected outcomes for the stated scenario under the stated assumptions. They are not general claims about charitable remainder trusts and do not transfer to a different client.
What drives the outcome
Four variables carry most of the result, and two of them interact strongly enough that screening on either one alone misclassifies clients near the boundary.
Asset basis is the gate
The trust's core economic advantage is shielding embedded gain. As basis rises the gain shrinks and the advantage disappears. In the published analysis, at 20 percent portfolio turnover the trust is projected to beat all three benchmarks where basis is below roughly 11 percent of value, and to lose to all three where basis exceeds roughly 25 percent. High basis effectively forecloses the case regardless of every other factor.
Longevity is the amplifier
Planners are trained to treat long life as a risk. Here the relationship inverts. The deduction is locked at inception on average life expectancy, but every additional year of survival delivers another year of tax-sheltered compounding and distributions while the taxable alternative absorbs another year of turnover drag. For the same baseline couple, a seven-year longevity adjustment moves the projected win probability against hold-and-draw from 28.2 percent to 96.4 percent.
That matters because the prescribed table is population-average, and the people who fund these trusts are not a population-average group. The income-longevity gradient at the top of the wealth distribution is steep and well documented.
Portfolio turnover and state tax
Both work through the same mechanism: they raise the annual tax drag on the taxable alternative. At 20 percent turnover, the same baseline case is projected to win 5.0 percent of the time in a zero-tax state versus 27.7 percent at California's 9.3 percent bracket. A recommendation calibrated to California materially overstates viability for a client in Texas or Florida.
Payout rate, and a structuring convention worth questioning
Charity-managed programs commonly steer donors to the 5 percent statutory minimum on the reasoning that preserving corpus lets the trust grow. That is coherent if the client's objective is growing nominal income over time. If the objective is present-value wealth, the analysis runs the other way: distributions accelerated into early years are discounted less, and that front-loading outweighs the smaller deduction a higher payout produces. On the published figures, defaulting to the minimum rather than setting the rate near the 10 percent remainder test ceiling costs roughly six percentage points of projected win probability.
A lower payout also produces a larger charitable remainder. That is not a criticism — the interests need not conflict — but the tradeoff should be made deliberately rather than absorbed as a default.
How many of these exist
The charitable remainder trust literature has been citing a 2014 IRS study of filing year 2012 for a decade. Analysis of IRS Form 5227 microdata through filing year 2022 updates the picture:
- 95,165 charitable remainder trusts in 2022, down from 105,860 in 2012 — a decline of about 10 percent, or negative 1.06 percent compounded annually.
- 74,552 unitrusts and 20,613 annuity trusts. Annuity trusts grew in absolute number over the decade while unitrusts fell.
- About 2,462 new trusts formed per year across 2015–2020 — roughly 17 percent above the rate implied by the 2012 study, contrary to the expectation that the larger estate tax exemption would suppress formations.
- Mean age of an active trust: 18.2 years; median 20. Terminations are outpacing formations, consistent with the aging of trusts created during the 1990s boom.
Two cautions on the same data. The asset-size fields in the microdata are unreliable — the reported distribution fails every standard model of asset-size data, and the largest single reported entry alone would exceed the total assets of all such trusts nationally. Population counts and type classifications are sound; asset figures are not. And there is no reliable per-trust median value in any public source. The only defensible size figure is an average of roughly $900,000, derived from the 2012 aggregate of $85.2 billion across 91,244 unitrusts.
Klaus Gottlieb, “Charitable Remainder Trusts, a Decade After the Last IRS Study,” Tax Notes Federal 190 (Mar. 9, 2026): 1613.
Where it works, and where it doesn't
Reduced to a screen, the evidence points in a consistent direction.
Conditions that favor the trust
- Deeply appreciated asset — basis low as a fraction of value
- Beneficiaries in good health, with reason to expect above-table longevity
- High-tax domicile
- An alternative portfolio that would be actively traded, generating annual realized gain
- A client whose realistic alternative is to sell and diversify, not to hold
- Genuine charitable intent, which converts the mandatory remainder from a cost into an objective
Conditions that argue against it
- Basis above roughly a quarter of value
- Older beneficiaries with a short expected horizon
- Zero-tax or preferential-rate state
- A buy-and-hold alternative with little annual turnover
- A client who would otherwise hold the asset to death and pass it with a stepped-up basis
- Any meaningful likelihood the client will need the principal back
That last one is not on the same footing as the others. Every quantitative comparison here scores present-value wealth and assigns nothing to the optionality the donor gives up permanently. A client who might need principal is not a marginal case that better modeling can resolve. Irrevocability is the answer to that question, and it is a clinical conversation, not a calculation.
Running the numbers on a specific client
Two free calculators handle the deduction side. The CRUT deduction calculator applies the IRS actuarial methodology and the current section 7520 rate across the full input range, and the payout path tool illustrates distribution paths under assumed returns. They compute the deduction. They do not answer whether the trust beats the alternative.
QuantiCRUT™ is the framework described above, implemented as software: 10,000 simulated paths, all three benchmarks, both a population and an annuitant mortality basis, with the projected win probability and the median dollar difference for each. The output is a report an advisor can put their own name on.
Plans from $325. Nothing stored — inputs exist only to render the PDF. Sample reports and the underlying research.
Common questions
Is income from a charitable remainder trust tax-free?
No. The trust pays no tax on the sale of appreciated property, but distributions are taxable to the beneficiary under the four-tier system of section 664(b). Ordinary income comes out first, then capital gain — including the entire gain realized when the trust sold the contributed asset — then tax-exempt income, then tax-free principal. Because gain accumulates ahead of principal in the ordering, most beneficiaries never receive a tax-free distribution.
Can I change my mind after funding a charitable remainder trust?
No. The trust is irrevocable. The donor cannot reach principal, cannot unwind the structure, and cannot recover the property. Limited options exist — the income interest can sometimes be sold or given to the charity to accelerate termination, and some states permit judicial modification in narrow circumstances — but none of them return the asset to the donor on the original terms.
What is the 10 percent remainder test?
The present value of the charitable remainder, computed at funding using the section 7520 rate and the prescribed mortality table, must be at least 10 percent of the initial fair market value of the contributed property. Sections 664(d)(1)(D) and 664(d)(2)(D). The test caps how high a payout rate can be set for a given age profile: younger beneficiaries and longer terms leave less remainder, so they must accept a lower rate to qualify.
Does the section 7520 rate matter for a unitrust?
Less than commonly assumed. In the unitrust remainder computation the rate serves as both the assumed growth rate of the corpus and the discount rate for beneficiary distributions, and the two effects cancel, leaving the remainder factor a function of the payout rate and mortality. The rate still governs whether a given payout-and-age combination clears the 10 percent test. In simulation, sweeping the rate from 1.2 percent to 8.2 percent shifts projected win probability by fewer than five percentage points. Timing a trust around IRS rate publications is tuning a variable that barely moves the outcome.
Does a charitable remainder trust avoid estate tax?
Usually there is no estate tax, but not because the assets are excluded. A donor who retains a lifetime interest causes the corpus to be included in the gross estate under section 2036(a)(1). The charitable remainder then qualifies for the estate tax charitable deduction under section 2055, and where the surviving spouse is the sole other beneficiary the spousal interest qualifies for the marital deduction under section 2056(b)(8). Naming a non-spouse successor beneficiary changes the analysis and generally creates a completed gift at funding unless the donor retains the right to revoke that interest by will.
Are the assets protected from creditors?
The corpus is beyond the donor's reach and generally beyond a creditor's. The donor's retained income interest is not. A self-settled spendthrift restriction on a retained interest is ineffective against the donor's creditors in most states, so a creditor who cannot attach the trust assets can usually attach the payment stream.
What payout rate should I choose?
It depends on the objective, and the two objectives point in opposite directions. A lower rate preserves corpus and grows nominal income over the trust's life. A higher rate maximizes present-value wealth, because accelerated distributions are discounted less and that gain outweighs the smaller deduction. Charity-managed programs commonly recommend the 5 percent minimum. On the published present-value analysis, defaulting to the minimum rather than setting the rate near the 10 percent test ceiling costs roughly six percentage points of projected win probability. Make the choice deliberately.
Can I fund a charitable remainder trust with mortgaged real estate?
It is a poor idea without clearing the debt first. Debt-financed income is unrelated business taxable income under section 514, and a charitable remainder trust with any UBTI in a year pays a 100 percent excise tax on it under section 664(c). Encumbered property also raises grantor trust exposure under section 677 and self-dealing questions where the donor remains personally liable.
How much does it take to make a charitable remainder trust worthwhile?
There is no statutory minimum, and the frequently quoted $200,000 threshold has no authority behind it. The real constraint is that drafting, appraisal, annual accounting, tax filing, and trustee fees are largely fixed, so they consume a larger share of a small trust. The more useful question is not the dollar size but whether the basis, ages, domicile, and realistic alternative make the structure work at all — a $5 million trust funded with high-basis property can be the wrong answer while a smaller one funded with a near-zero-basis position is the right one.
Do charitable remainder trusts usually outperform the alternative?
Not usually, on present-value wealth alone, and it depends heavily on which alternative is used. Across 500 simulated client profiles spanning a realistic parameter range, the trust was projected to prevail against the strictest benchmark — hold and draw — in 34.4 percent of cases. Against the immediate-sale benchmark it prevails far more often. The instrument is highly situational rather than generally favorable or generally unfavorable, and the conditions that decide it are identifiable in advance.