ConceptsTax Strategy21 min readPublished September 2, 2026

Should Charitable Donations Be Tax Deductible? The Case for Ending the Deduction

The US charitable deduction costs about $69B a year and 63% of its 2019 benefit went to the top 1%. What the research says, and why I would end it anyway.

This is a policy argument. I think charities do a great deal of good and that the tax code should stop subsidizing gifts to them. The two positions are compatible. Figures are for the United States under the law in force for tax year 2026 unless stated otherwise.

The federal charitable deduction will cost about $69 billion in forgone income tax in fiscal 2026, by the Treasury’s own estimate.1 In 2019, the last year the Congressional Budget Office scored its distribution, 93% of that benefit went to households in the top fifth of the income distribution and 63% to the top 1%.2 The best recent evidence on what the subsidy buys is a 2024 study of the 2017 tax reform, which removed the giving incentive for about a fifth of taxpayers. It found that a 10% cut in the after-tax price of giving raises giving by about 6% for the average donor.3 At that response, the Treasury gives up more than the charities receive, and most of the subsidy pays for gifts that would have been made anyway.

The Economist made the case for scrapping these breaks in August 2026, and it is a case I agree with.4 My position: end the individual income-tax deduction, including the fair-market-value treatment of appreciated property, over a multi-year phase-down. Keep charities tax-exempt. If Congress insists on subsidizing gifts, do it with a flat, capped credit paid as a match to the charity, which the evidence says buys more giving per dollar of revenue than a deduction claimed on a tax return.

What “tax deductible” means, and what repeal would touch

A common reply to any proposal like this is that charities do not make a profit, so there is nothing to tax. That confuses two provisions. A charity’s tax-exempt status means the organization itself owes no income tax on its receipts. The charitable deduction is a separate rule that lets the donor subtract the gift from their own taxable income. Ending the second does nothing to the first. Repeal proposals also need to say which of several related provisions they cover, because each is a separate decision.

ProvisionWhat it doesTouched by repealing the deduction?
Individual income-tax deductionDonor subtracts the gift from taxable incomeYes. This is the $69 billion item.
Charity’s tax-exempt statusQualifying organizations pay no income taxNo.
Appreciated-property treatmentDeduct full market value; the unrealized gain is never taxedOnly if the proposal says so. I would include it.
Qualified charitable distributionsIRA owners 70½ and older exclude up to $111,000 (2026) of direct gifts from incomeNo. It is an exclusion, and a separate vote.
Estate-tax deductionBequests to charity leave the taxable estateNo.
Corporate deductionCompanies deduct gifts above a 1% floor, up to 10% of taxable incomeNo.

The 2026 rules from the One Big Beautiful Bill Act already changed the individual deduction in three ways. Non-itemizers can now deduct up to $1,000 ($2,000 on a joint return) of cash gifts, though not gifts to donor-advised funds. Itemizers can deduct only the part of their giving above 0.5% of adjusted gross income. And filers in the 37% bracket get at most 35 cents of tax benefit per dollar of itemized deductions.56 Those are real changes, and two of them push in the right direction. They leave the structure intact.

Where the deduction came from

The deduction entered the code in the War Revenue Act of 1917, the year the top income tax rate reached 67%, because senators worried that wealthy donors would cut gifts to colleges and the Red Cross rather than cut their own spending. Senator Henry Hollis of New Hampshire argued that “when war comes and we impose these very heavy taxes on income, that will be the first place where wealthy men will be tempted to economize, namely in donations to charity.” The deduction was capped at 15% of income.7

The rest of the history is a series of patches. A corporate deduction followed in 1935. The standard deduction arrived in 1944 and turned the charitable deduction into an itemizer’s benefit, which it has been ever since. In the 1950s and 1960s, high marginal rates turned industrial fortunes into large private foundations, which set off a national argument about whether founders were buying permanent control of tax-favored wealth. The Tax Reform Act of 1969 answered with payout rules, self-dealing prohibitions, and excise taxes on foundations, and the 5% minimum distribution rule that applies today descends from it.8 Historian Nicolas Duquette describes that decade as the point when lawmakers stopped describing philanthropy as private wealth substituting for public spending and started treating the deduction as “an implicit public cost.”9

Congress has tried extending the deduction to non-itemizers before. The 1981 tax act phased one in from 1982 and let it expire after 1986. A Treasury official at the time, Donald Lubick, objected that an above-the-line deduction “would go, in very large measure, to those who are already giving with respect to their existing gifts.”7 That objection is the core of the modern case, and it applies to the itemized deduction as well.

The test the deduction has to pass

For an itemizing donor whose marginal tax rate is t, a dollar given to charity reduces tax by t, so the after-tax price of giving a dollar is 1 − t. In the 32% bracket a $10,000 gift costs the donor $6,800 and costs other taxpayers $3,200. The subsidy applies to the whole gift, including the part the donor would have given at full price. If that donor would have given $8,000 with no deduction and gives $10,000 with it, the Treasury has paid $3,200 to move $2,000 to the charity. The rest of the subsidy is a transfer to the donor.

Economists summarize the response with a price elasticity: the percentage change in giving for a 1% change in its after-tax price. Under the conventional test, a subsidy pays for itself only if the elasticity exceeds 1 in absolute value, because that is the point at which the extra giving it causes equals the revenue it costs. Below 1, charities gain less than the Treasury loses.10 The calculator below splits one gift along those lines and lets you set the elasticity, since that is the number the whole debate turns on.

Two qualifications belong next to that test. Emmanuel Saez showed in 2004 that the optimal subsidy depends on the elasticity, on how much private giving crowds out public provision, and on the size of the public benefit from the funded activity, and his simulations put the optimal subsidy rate below the income tax rate in most cases, which is to say below full deductibility.11 Louis Kaplow goes further: once the deduction sits inside an optimal-tax model, where the government can adjust rates elsewhere to offset revenue and distributional effects, the elasticity “does not merely fail to be a sufficient statistic for welfare analysis; it is not even a relevant one.” What matters is the net external benefit of the marginal gift.12 So an elasticity below 1 is strong evidence that the deduction buys little giving per dollar of revenue. It does not by itself prove that the right subsidy is zero, and I will come back to that.

What the research finds

The literature does not deliver a single elasticity. It delivers a range, and where a study lands depends on the data, the country, and how it separates the effect of the tax price from the effect of income, which move together in a progressive tax system.

StudySettingPrice elasticity (absolute value)Note
Randolph (1995)US tax-return panel, 1979 to 19880.5 permanent; 1.55 transitoryDonors time gifts around rate changes; permanent response is small
Auten, Sieg & Clotfelter (2002)US tax-return panel, 1980 to 19920.79 to 1.26 persistentPersistent price changes matter more than transitory ones
Bakija & Heim (2011)US tax-return panel, 1979 to 20061.4 (confidence interval 1.0 to 1.8)Identified from differences in state tax paths; supports treasury efficiency
Fack & Landais (2010)French tax-credit reforms0.2 to 0.6 across gift sizesExtra giving was smaller than forgone revenue; response rises with gift size
Almunia, Guceri, Lockwood & Scharf (2020)UK self-assessment returns, 2005 to 20130.3 total (0.2 intensive, 0.1 extensive)Their welfare model still found a case for a larger UK subsidy
Duquette (2016)US charity-level receipts around the 1986 reformAbout 4Measures charity revenue, not donor deductions; the outlier
Meer & Priday (2020)US household panel, 2001 to 20171.07 overall; 0.79 for continuing itemizersProjected the 2017 reform would cut giving by about $9.8 billion
Han, Hungerman & Ottoni-Wilhelm (2024)US, the 2017 reform as a natural experiment0.6 average; over 2 for the most responsive donorsGiving fell about $20 billion a year; almost none of the drop was in gifts to congregations

Sources: references 3, 13 to 19 in the list at the end. Randolph’s figures are as reported by Auten, Sieg and Clotfelter (2002).

The 2024 paper by Xiao Han, Daniel Hungerman, and Mark Ottoni-Wilhelm deserves the most weight, because the 2017 reform was the cleanest experiment the United States has run. By roughly doubling the standard deduction, it removed the federal giving incentive for about 20% of income-tax payers while leaving it in place for others with similar incomes. The share of returns that itemized fell from 30.6% in 2017 to 11.4% in 2018.20 Comparing the two groups, and adjusting for gifts pulled forward into 2017 by donors who saw the change coming, the authors estimate the reform cut giving by about $20 billion a year, with a permanent price elasticity of about 0.6 for the average donor and above 2 for the donors their model predicted would respond most.3

That result rules out two lazy positions at once. Incentives do affect giving; $20 billion a year is not nothing. And the average incentive does not pay for itself; at 0.6, the Treasury gives up more than charities gain. The spread between 0.6 and 2 also matters. A blunt repeal removes both the inframarginal subsidy on gifts that would happen anyway and a productive incentive for a minority of donors, and the average elasticity hides where the induced gifts come from.

Two further findings sharpen the picture. The same study found that the 2017 reform caused little change in giving to religious congregations; the decline was almost entirely in gifts to other organizations, which suggests congregational giving behaves more like dues than like a price-sensitive purchase.3 And an older paper by Jonathan Gruber found that larger subsidies to giving raise religious giving but reduce religious attendance, with people substituting money for time.21 Religion received $151.6 billion of the $617.2 billion Americans gave in 2025, the largest single category.22

Why the estimates differ, and why design matters

Tax rates rise with income, so separating the price effect from the income effect is hard. Reforms change several provisions at once. Tax returns record deductions claimed rather than gifts made, and Denmark’s switch to third-party reporting of donations in 2008 doubled the number of deductions claimed without any evidence of a change in giving.23 Donors shift gifts across years to exploit anticipated rate changes, which inflates short-run estimates relative to permanent ones. And donor-level and charity-level data measure different things, which is part of why Duquette’s estimate sits so far from the others.

The form of the subsidy matters as much as its size. A donation match, where the charity receives the government’s contribution directly, produces far more giving per dollar than a rebate claimed by the donor months later. In a study of gifts to an Indiana university, Hungerman and Ottoni-Wilhelm estimated a match-price elasticity of about 1.2 against a rebate-price elasticity of about 0.2.24 Survey evidence from UK Gift Aid donors points the same way, mostly because a majority of donors do not adjust their nominal gift when the rebate changes.25 The US deduction is the weakest design on this axis: a rebate, claimed on a return filed the following spring, worth a different amount to every donor.

Who receives the tax benefit

A deduction is worth the donor’s marginal rate. The same $1,000 gift saves a 37%-bracket filer up to $350 and a 12%-bracket filer $120, and saves a non-itemizer nothing beyond the new $1,000 or $2,000 allowance. Tax analysts call this an upside-down subsidy, and the distribution follows from it.26

  • CBO, 2019: of a $43 billion tax expenditure, 93% of the benefit went to the top income quintile and 63% to the top 1%. Only 11% of filers itemized.2
  • Joint Committee on Taxation, 2025 income levels: returns with expanded income above $1 million received 61.5% of the deduction’s tax expenditure; returns above $500,000 received 73.6%; returns above $200,000 received 92.1%.27
  • IRS Statistics of Income, tax year 2020: returns with adjusted gross income of $500,000 or more claimed 56% of all itemized charitable deduction dollars, on 8.6% of the returns that claimed one.28 That matches The Economist’s “more than half” figure, and it understates the concentration of the tax benefit, since those returns also face the highest rates.
  • Tax Policy Center: the roughly 10% of households that itemize give about 65% of individual giving dollars, and the average federal subsidy across all giving is about 17%, down from about 21% before the 2017 reform.29

Two qualifications. These figures describe who receives the tax reduction, not who benefits from the funded services, which may be very different people. And they describe pre-2026 law. The new non-itemizer allowance spreads a small benefit widely, and the 35% cap and 0.5% floor trim the top. Neither changes the fact that the subsidy rate still rises with income.

The “taxes as ballots” defense, and where it fails

The most interesting argument for the deduction is that it decentralizes a spending decision. Saul Levmore put it this way in a 1998 article: each donor’s deduction is a ballot that triggers a matching government contribution, so the public directs part of the budget toward causes it knows and cares about, rather than leaving the whole allocation to legislators.30 Donors do have information the state lacks, and a deduction lets minority preferences, local needs, and unpopular experiments get funded without winning a majority vote.

Levmore himself noted the two objections that undo it. You have to pay to vote, since only a taxpayer with a liability can cast a ballot. And a deduction, unlike a credit, gives more votes to higher-bracket taxpayers.30 Rob Reich’s book on the subject calls the result a policy that “bakes a plutocratic bias into the incentive to give, systematically benefiting the wealthy and amplifying their voices.”31 A voting system in which one household holds hundreds of thousands of ballots and most households hold none is a strange thing to defend as democratic.

A third problem is that the ballot never shows the cost. When a donor decides to give, the cause is in front of them and the public services that will get a few dollars less are not. Legislatures, for all their faults, have to weigh spending priorities against each other and against taxes, in public, with a vote that can be reversed. Tax expenditures skip that process entirely: they are not appropriated, they are not voted on annually, and their cost appears in a Treasury appendix rather than in a budget line.1 And the state must remain neutral about which registered charities deserve support, which is the right rule for the state and a poor rule for a subsidy. Every 501(c)(3) qualifies alike, whether it runs a food bank, a donor-named building, an advocacy shop, a private art gallery, or a congregation, and churches are not required to apply for recognition or file the annual Form 990 that other charities file.32

Where deducted money sits

The tax expenditure counts dollars deducted, which is not the same as dollars reaching an operating charity in that year. Three structures separate the two.

Donor-advised funds. A donor contributes to a sponsoring charity, takes the deduction immediately, gives up legal control, and keeps advisory privileges over investment and grants.33 At the end of fiscal 2024, US donor-advised funds held $327.9 billion, took in $90.6 billion, and granted out $64.6 billion, an aggregate payout of about 25% of prior-year assets.34 That average is healthy. It is an average, and there is no legal minimum payout for an individual account; the Government Accountability Office flagged the reporting gap in 2006, and IRS regulations proposed in 2023 remain proposed.35 For an investor, a DAF is a useful tool. As a matter of public accounting, the deduction was claimed the year the money went in, and the charity may be waiting.

Private foundations. Foundations must distribute about 5% of investment assets each year, a floor set after the 1969 debate.8 The Musk Foundation shows how little that floor requires. Its assets had a fair market value of about $14.7 billion at the end of 2024. It gave out $474 million that year, which The New York Times reported was $393 million short of the minimum, the fourth consecutive year it fell short. Its largest 2024 grant, $370 million, went to a nonprofit run by Musk’s wealth manager that operates a school and child-care program in Bastrop, Texas, near his companies.36 The deductions for the gifts that built that endowment were taken years ago.

Private museums. Sheldon Solow, a New York property developer who died in 2020, gave works by Botticelli, Matisse, and Miró to his own foundation over two decades. Dealers estimated the collection at around $500 million; the foundation’s own filings carry it at roughly $218 million at book value. Through 2020 the foundation’s charitable disbursements ran under $350,000 a year, and the ground-floor gallery on West 57th Street was closed to the public.37 It opened in 2023 under his widow and now offers open viewing on Thursday afternoons plus tours by appointment. The Senate Finance Committee had reviewed private museums in 2016 and called them “ripe for exploitation.”37

Appreciated property. A donor who gives long-held stock deducts its full market value and never pays tax on the gain. That is two subsidies stacked on one gift, and the second is the larger one for a concentrated-stock holder. Even William Andrews, whose 1972 article is the classic defense of the cash deduction on tax-base grounds, wrote that the fair-market-value rule “must now be viewed as a subsidy or artificial inducement, above and beyond mere tax exemption, for philanthropic giving.”38 The mechanics are in Donate Appreciated Stock or Use a DAF?.

Would the government spend the money better?

The case for repeal does not need that claim, and I would not make it. Government grants to charities crowd out private giving: in a panel of more than 8,000 charities, James Andreoni and Abigail Payne found that a $1,000 government grant reduced private contributions by about $727, with the whole effect running through charities cutting their own fundraising once the grant arrived.39 Public programs can be slow, majoritarian, and bad at experimentation. Charities can be duplicative, donor-driven, and unaccountable. They are two imperfect institutions that partly substitute for each other.

The argument rests on something narrower. Every dollar of tax expenditure has an opportunity cost, whether it is lower rates, lower deficits, or public spending. Appropriations pass through a budget, an audit, disclosure rules, and an election. The deduction passes through none of them, and it delegates the allocation of public money to whoever has the largest tax bill. A country that wants to fund food banks and medical research can do so directly, on the record, at a cost it can see.

The strongest case for keeping it

Here is the other side at full strength.

  • Public goods are underprovided by markets. Medical research, poverty relief, and open-access culture benefit people who never donate, so voluntary giving falls short of the efficient level, and a subsidy is the textbook fix. Kaplow’s 1995 note adds that even pure altruistic gifts carry an externality, because the donor does not count the recipient’s gain as society would.40
  • The tax-base argument. Andrews argued that income given away is not consumed by the donor and so should not be in the donor’s tax base at all; the deduction defines income correctly rather than subsidizing anything.38 The reply is that the donor does consume something: standing, naming rights, membership, and the satisfaction of directing the gift. But the argument is serious.
  • Pluralism. Without a subsidy, more of civil society’s funding decisions move to legislatures, which fund median-voter causes. Some of what the deduction pays for, from unpopular research to minority religious communities, would not survive a majority vote and should not have to.
  • Repeal would cut giving. The 2024 evidence says losing the incentive cost about $20 billion a year for a fifth of taxpayers.3 Full repeal would cost more, and food banks, hospitals, and research institutes have built budgets around it. That is a transition cost, and it is not small.
  • The welfare math can go either way. Almunia and co-authors measured a UK elasticity of 0.3 and still concluded that the UK subsidy should be larger, because their model valued the funded public goods highly.15 Saez’s framework gives a positive optimal subsidy in most simulations, just one below the tax rate.11

The last point is the one I take most seriously. The research establishes that the current design buys little giving per revenue dollar, has an upside-down distribution, and no mechanism to distinguish a food bank from a founder’s art vault. It does not establish that the optimal subsidy is exactly zero. My preference for zero rests on the governance and equal-treatment arguments above, plus a simplicity argument: a tax code with one fewer itemized deduction is a better tax code. Those are value judgments, and I hold them with the evidence in view rather than because the evidence compels them.

What I would do instead

The alternatives fall on a spectrum, and each one fixes a different defect.

DesignWhat it fixesWhat it costs
Repeal the individual deduction and the appreciated-property ruleEnds the upside-down subsidy, the donor-directed budget, and the double benefit on stockGiving falls; a phase-down is needed for organizations built around it
Keep the deduction, add a floorStops subsidizing the first dollars, which are the most likely to be inframarginalKeeps the income-linked subsidy above the floor. The 2026 law did a small version of this
Flat credit for all filersSame subsidy rate at every income; Canada, France, and New Zealand use creditsRate, cap, and refundability decide the cost; a subsidy still exists
Match paid to the charity (UK Gift Aid)Salient at the moment of giving; buys more giving per revenue dollar than a rebateStill delegates public spending to donors; the UK version keeps a higher-rate rebate on top
Cause-specific subsidyTargets activities with measured public benefitRequires the state to rank charities, and invites lobbying over the ranking
DAF and foundation payout reformCloses the gap between the year of the deduction and the year of the grantLeaves the deduction itself untouched

CBO simulated most of these against 2006 law. In every option it examined that added a floor, the reduction in the subsidy exceeded the reduction in contributions. A 2%-of-AGI floor for itemizers cut giving by $3.0 billion and the subsidy by $15.7 billion. Extending the deduction to all filers with no floor, the direction Congress moved in 2026, added $2.0 billion of giving at a cost of $5.2 billion.10 Those are old simulations, and the mechanism has not changed: floors remove the subsidy on gifts that were coming anyway, and universal deductions without floors add subsidy faster than giving.

My recommendation, in order of preference:

  1. Phase out the individual income-tax deduction over four years, and end the fair-market-value rule for appreciated property with it. Keep tax-exempt status for charities, keep the estate-tax deduction and qualified charitable distributions as separate questions for separate votes, and put the recovered revenue on the record where it can be appropriated or returned in lower rates. The phase-down is the transition cost the 2024 evidence says the sector needs.
  2. If a subsidy survives, make it a flat, capped credit paid as a match to the charity. A uniform credit is the efficient form of a tax incentive when there is no evidence that the public benefit of a gift rises with the donor’s income, which there is not.41 Paying it as a match rather than a rebate roughly doubles what each revenue dollar buys, on the Indiana evidence. The OECD, surveying 40 countries in 2020, recommended credits over deductions for exactly the distributional reason.42
  3. Either way, require a minimum payout from donor-advised funds and enforce the foundation minimum. A deduction claimed in a year when no charity received anything is the clearest defect in the system, and it is fixable without touching the larger question.

What this means for your own giving

None of this changes the first-order rule: decide whether and where to give based on what you want to accomplish and how effective the organization is, then optimize the tax mechanics. A deduction never makes you richer for giving. It reduces the amount you sacrifice, by at most 35 cents on the dollar at the federal level in 2026, and the other 65 cents is gone.

  • Treat the deduction as a policy at risk. The 2026 law already capped it at the top and floored it at the bottom, and the direction of every serious reform proposal in the last fifteen years has been to shrink it. Do not build a multi-decade giving plan that only works at today’s subsidy rate.
  • Under current law, appreciated stock beats cash. Giving long-held winners avoids the capital-gains tax and takes the deduction on full market value. This is the part of the subsidy I would repeal, and the part I use while it exists. The after-tax cost calculator is in the appreciated stock and DAF guide.
  • Bunch gifts to clear the floor and the standard deduction. With a 0.5%-of-AGI floor and a standard deduction most households do not exceed, concentrating two or three years of giving into one year, often through a DAF, is what turns a gift into a deduction at all. Then grant from the DAF on your own schedule.
  • Over 70½, use qualified charitable distributions first. Up to $111,000 in 2026 can go directly from an IRA to a charity, excluded from income rather than deducted, which means it works whether or not you itemize and keeps the money out of the calculations that set Medicare premiums and Social Security taxation.43
  • Model the whole tax year. The value of a deduction depends on your bracket, your other itemized deductions, and the 35% cap. Run the year with and without the gift before deciding on timing.

See what a gift does to this year's tax bill

Summitward's tax projection runs your federal and state brackets year by year, so you can compare bunching two years of giving into one against spreading it out.

Open the tax projection

Key takeaways

  • Being pro-charity and anti-deduction is a coherent position. Ending the donor’s deduction leaves charities tax-exempt. The question is whether a $69 billion subsidy, delivered through the tax code and directed by donors, is a good way to fund public goods.
  • The subsidy buys little extra giving. The best US evidence, from the 2017 reform, puts the average price elasticity at 0.6, so the Treasury loses more than charities gain, and at a 30% marginal rate roughly four fifths of subsidized giving would have happened without the subsidy under a constant-elasticity model.
  • The benefit is concentrated. In 2019, 63% of the deduction’s tax benefit went to the top 1% of households and 93% to the top fifth. A deduction pays more per dollar to whoever has the highest rate.
  • The design is the weakest available. Rebates claimed on a return the following year buy far less giving than matches paid to the charity, and the deduction distinguishes a food bank from a private art gallery not at all.
  • The evidence supports shrinking the subsidy more than it settles the optimal level. Zero is my position on governance and equal-treatment grounds; a flat credit paid as a match is the defensible second-best.
  • For your own giving, decide the cause first and the tax second, use appreciated stock, bunching, and qualified charitable distributions while they exist, and do not assume they will.

Frequently asked questions

Should charitable donations be tax deductible?

The evidence says the current US deduction is an expensive way to encourage giving: a 10% cut in the price of giving raises giving about 6% for the average donor, and most of the tax benefit goes to the highest-income households. Whether any subsidy should survive is a value judgment about who should direct public money. My view is that it should not, and that a flat credit paid as a match to the charity is the acceptable compromise.

Would ending the deduction mean taxing charities?

No. A charity’s exemption from income tax and the donor’s deduction are separate provisions. Repealing the deduction changes what the donor owes, and nothing about what the charity owes.

How much does the charitable deduction cost?

The Treasury estimates the individual deduction at about $68.9 billion in fiscal 2026 and $69.4 billion in fiscal 2027. The Joint Committee on Taxation puts fiscal 2026 at $72.9 billion for individuals. Neither is a repeal revenue estimate, because both hold behavior and other provisions constant.

Who benefits most from the charitable deduction?

High-income itemizers. The Congressional Budget Office found 63% of the benefit went to the top 1% of households in 2019, and the Joint Committee on Taxation estimates that returns with income above $500,000 receive about 74% of the tax expenditure at 2025 income levels.

Do tax deductions actually increase charitable giving?

Yes, and by less than they cost at the average. When the 2017 reform removed the incentive for about a fifth of taxpayers, giving fell by about $20 billion a year. The estimated response, about 0.6, means the extra giving was smaller than the forgone revenue. Some donors respond much more strongly, and the average hides them.

What changed for charitable deductions in 2026?

Non-itemizers can deduct up to $1,000 ($2,000 joint) of cash gifts, excluding gifts to donor-advised funds. Itemizers can deduct only giving above 0.5% of adjusted gross income. Filers in the 37% bracket get at most 35 cents of benefit per dollar of itemized deductions. The 60%-of-AGI limit for cash gifts is now permanent.

What would replace the deduction?

The main candidates are a flat credit at the same rate for every filer, which Canada, France, and New Zealand use, or a match paid directly to the charity, as in the UK’s Gift Aid. A uniform credit removes the income link, and a match buys more giving per revenue dollar than a rebate.

Related guides

Sources

  1. US Department of the Treasury, Office of Tax Analysis, “Tax Expenditures,” fiscal year 2027 report (December 16, 2025), Table 1; charitable contributions to education ($9,600 million), health ($7,610 million), and other ($51,650 million) for fiscal 2026, individuals. The report states that its estimates “do not necessarily equal the increase in federal receipts” from repeal. home.treasury.gov
  2. Congressional Budget Office, “The Distribution of Major Tax Expenditures in 2019” (October 2021), Table 2 and text: $43 billion; “Ninety-three percent of its benefits accrued to households in the highest quintile, including 63 percent to households in the top 1 percent”; 11% of filers itemized. cbo.gov
  3. Xiao Han, Daniel M. Hungerman, and Mark Ottoni-Wilhelm, “Tax Incentives for Charitable Giving: New Findings from the TCJA,” NBER Working Paper 32737 (July 2024): incentives eliminated for “roughly 20 percent of US income-tax payers”; giving down “about $20 billion annually”; permanent price elasticity “from .6 for the average donor to over 2 for those predicted to be most responsive”; “little change in giving to religious congregations.” nber.org
  4. The Economist, “Tax breaks for charity donations should be scrapped,” Finance & economics (August 27, 2026). economist.com
  5. Internal Revenue Service, Topic No. 506, “Charitable contributions”: “Beginning with tax year 2026, if you do not itemize, you may deduct up to $1,000 ($2,000 if filing jointly) of your cash contributions.” The exclusion of donor-advised funds is in new IRC section 170(p). irs.gov
  6. Senate Committee on Finance, section-by-section summary of Title VII of Public Law 119-21 (2025): sec. 70111 (itemized deductions capped at $0.35 per dollar for the 37% bracket, IRC section 68), sec. 70424 (non-itemizer deduction), sec. 70425 (0.5% floor, IRC section 170(b)(1)(I), and permanent 60% limit), sec. 70426 (1% corporate floor). finance.senate.gov
  7. Margot L. Crandall-Hollick, “The Charitable Deduction for Individuals: A Brief Legislative History,” Congressional Research Service Report R46178 (updated June 26, 2020); Hollis quotation from Congressional Record vol. 55 (September 7, 1917), p. 6728; Lubick quotation from 1981 testimony. congress.gov
  8. Internal Revenue Service, “Minimum investment return” (private foundations): 5% of the fair market value of non-exempt-use assets; IRC section 4942, enacted by the Tax Reform Act of 1969. irs.gov
  9. Nicolas J. Duquette, “Founders’ Fortunes and Philanthropy: A History of the U.S. Charitable-Contribution Deduction,” Business History Review 93(3), 2019, pp. 553-584. cambridge.org
  10. Congressional Budget Office, “Options for Changing the Tax Treatment of Charitable Giving” (May 2011), and Frank Sammartino’s testimony before the Senate Finance Committee (October 18, 2011), Table 1: 2006 baseline of $203.0 billion in contributions and a $40.9 billion subsidy; “In each case that CBO examined, the reduction in the subsidy ... would exceed the reduction in charitable contributions.” cbo.gov
  11. Emmanuel Saez, “The optimal treatment of tax expenditures,” Journal of Public Economics 88(12), 2004, pp. 2657-2684: the optimal subsidy “in most cases ... should be lower than the earnings tax rate.” eml.berkeley.edu
  12. Louis Kaplow, “Optimal Income Taxation and Charitable Giving,” Tax Policy and the Economy 38 (2024). law.nyu.edu
  13. William C. Randolph, “Dynamic Income, Progressive Taxes, and the Timing of Charitable Contributions,” Journal of Political Economy 103(4), 1995, pp. 709-738; permanent price elasticity of -0.51 and transitory -1.55 as reported in Auten, Sieg and Clotfelter (2002). ideas.repec.org
  14. Gerald E. Auten, Holger Sieg, and Charles T. Clotfelter, “Charitable Giving, Income, and Taxes: An Analysis of Panel Data,” American Economic Review 92(1), 2002, pp. 371-382; persistent price elasticities “range from -0.79 to -1.26.” aeaweb.org
  15. Miguel Almunia, Irem Guceri, Ben Lockwood, and Kimberley Scharf, “More giving or more givers? The effects of tax incentives on charitable donations in the UK,” Journal of Public Economics 183, 2020: intensive-margin elasticity about -0.2, extensive -0.1, total about -0.3; “there is a case for increasing the subsidy on charitable giving in the UK.” wrap.warwick.ac.uk
  16. Jon Bakija and Bradley T. Heim, “How Does Charitable Giving Respond to Incentives and Income? New Estimates from Panel Data,” National Tax Journal 64(2), 2011, pp. 615-650; state-identified persistent price elasticity of -1.4 (standard error 0.2). web.williams.edu
  17. Gabrielle Fack and Camille Landais, “Are Tax Incentives for Charitable Giving Efficient? Evidence from France,” American Economic Journal: Economic Policy 2(2), 2010, pp. 117-141; elasticities “around -0.2 to -0.6 across quantiles.” aeaweb.org
  18. Nicolas J. Duquette, “Do tax incentives affect charitable contributions? Evidence from public charities’ reported revenues,” Journal of Public Economics 137, 2016, pp. 51-69: “A one percent increase in the tax cost of giving causes charitable receipts to fall by about four percent.” ideas.repec.org
  19. Jonathan Meer and Benjamin A. Priday, “Tax Prices and Charitable Giving: Projected Changes in Donations under the 2017 Tax Cuts and Jobs Act,” Tax Policy and the Economy 34, 2020, pp. 113-138 (NBER Working Paper 26452): a 10% price increase “is expected to reduce giving by 10.7 percent”; -0.79 for continuing itemizers; projected reduction of about $9.8 billion. nber.org
  20. Internal Revenue Service, Statistics of Income, Individual Income Tax Returns, Table 1.2, tax years 2017 and 2018: 46.9 million of 152.9 million returns itemized in 2017 (30.6%); 17.5 million of 153.8 million in 2018 (11.4%); 15.1 million of 160.6 million in 2023 (9.4%). irs.gov
  21. Jonathan Gruber, “Pay or pray? The impact of charitable subsidies on religious attendance,” Journal of Public Economics 88(12), 2004, pp. 2635-2655; implied elasticity of attendance with respect to religious giving of -0.92. nber.org
  22. Giving USA 2026: The Annual Report on Philanthropy for the Year 2025 (June 23, 2026): total giving $617.20 billion; individuals $394.2 billion; religion $151.58 billion. givingusa.org
  23. Christian Gillitzer and Peer Ebbesen Skov, “The use of third-party information reporting for tax deductions: evidence and implications from charitable deductions in Denmark,” Oxford Economic Papers 70(3), 2018, pp. 892-916. ideas.repec.org
  24. Daniel M. Hungerman and Mark Ottoni-Wilhelm, “Impure Impact Giving: Theory and Evidence,” Journal of Political Economy 129(5), 2021, pp. 1553-1614 (NBER Working Paper 24940): match-price elasticity about -1.2, rebate-price elasticity about -0.2. nber.org
  25. Kimberley Scharf and Sarah Smith, “The price elasticity of charitable giving: does the form of tax relief matter?” International Tax and Public Finance 22(2), 2015, pp. 330-352. ideas.repec.org
  26. C. Eugene Steuerle et al., “The Charitable Deduction: Economics versus Politics,” Urban Institute (April 2013): “the charitable deduction is an upside-down subsidy in the sense that the tax benefit increases for taxpayers with higher incomes.” urban.org
  27. Joint Committee on Taxation, “Estimates of Federal Tax Expenditures for Fiscal Years 2025-2029,” JCX-45-25 (December 3, 2025), Table 1 (individuals, fiscal 2026: $72.9 billion across the three charitable rows) and Table 3 (distribution by expanded-income class at 2025 levels: $44,110 million of $71,733 million to returns above $1 million). jct.gov
  28. Internal Revenue Service, Statistics of Income, Individual Income Tax Returns, Table 2.1, tax year 2020: contributions deduction of $204.66 billion on 12.64 million returns, of which $114.60 billion on 1.09 million returns with AGI of $500,000 or more. irs.gov
  29. Tax Policy Center, Briefing Book, “How large are individual income tax incentives for charitable giving?” and “How did the TCJA affect incentives for charitable giving?” taxpolicycenter.org
  30. Saul Levmore, “Taxes as Ballots,” University of Chicago Law Review 65(2), 1998, p. 387; the poll-tax and bracket objections appear at pp. 405-406. chicagounbound.uchicago.edu
  31. Rob Reich, Just Giving: Why Philanthropy Is Failing Democracy and How It Can Do Better (Princeton University Press, 2018); quotation from Reich in Stanford Magazine (March 2020). stanfordmag.org
  32. Internal Revenue Service, “Churches, integrated auxiliaries, and conventions or associations of churches”: churches “are automatically considered tax exempt and are not required to apply for and obtain recognition of exempt status” and “are not required to file an annual return.” irs.gov
  33. Internal Revenue Service, “Donor-advised funds”: “Once the donor makes the contribution, the organization has legal control over it. However, the donor ... retains advisory privileges.” irs.gov
  34. DAF Research Collaborative, Annual DAF Report 2025 (fiscal 2024 data): assets $327.87 billion, contributions $90.57 billion, grants $64.60 billion, payout rate 25.2%. dafresearchcollaborative.org
  35. Government Accountability Office, “Tax-Exempt Organizations: Collecting More Data on Donor-Advised Funds and Supporting Organizations Could Help Address Compliance Challenges,” GAO-06-799 (July 2006); proposed regulations on donor-advised fund distributions, REG-142338-07, Federal Register (November 14, 2023). gao.gov
  36. Theodore Schleifer and David A. Fahrenthold, “Elon Musk’s Foundation Grows to $14 Billion, but Gives Little to Outsiders,” The New York Times (December 2, 2025): $474 million given in 2024, $393 million short of the minimum, fourth straight year short; $370 million to The Foundation in Bastrop, Texas. Fair market value of $14.73 billion at year-end 2024 from the foundation’s Form 990-PF. nytimes.com
  37. Crain’s New York Business, “Developer Sheldon Solow’s museum is off-limits to the public” (April 2018), including the Senate Finance Committee’s 2016 review of private museums; Solow Art and Architecture Foundation Forms 990-PF via ProPublica Nonprofit Explorer (EIN 13-3614971): total assets of $217.8 million at book value, fiscal 2024; Soloviev Foundation gallery hours. projects.propublica.org
  38. William D. Andrews, “Personal Deductions in an Ideal Income Tax,” Harvard Law Review 86, 1972, p. 309; the fair-market-value quotation is at p. 372. ncpl.law.nyu.edu (summary)
  39. James Andreoni and A. Abigail Payne, “Is crowding out due entirely to fundraising? Evidence from a panel of charities,” Journal of Public Economics 95(5-6), 2011, pp. 334-343: “every $1000 grant reduces giving by $727 ... In our preferred specification, all of the crowd-out is attributable to fundraising.” econweb.ucsd.edu
  40. Louis Kaplow, “A Note on Subsidizing Gifts,” Journal of Public Economics 58(3), 1995, pp. 469-477 (NBER Working Paper 4868). nber.org
  41. Lily L. Batchelder, Fred T. Goldberg Jr., and Peter R. Orszag, “Efficiency and Tax Incentives: The Case for Refundable Tax Credits,” Stanford Law Review 59(1), 2006, p. 23. stanfordlawreview.org
  42. OECD, Taxation and Philanthropy, OECD Tax Policy Studies No. 27 (2020): 22 of 40 surveyed countries use deductions, 12 use credits, five use matching; recommends countries “consider providing a tax credit for donations instead of a deduction.” oecd.org
  43. Internal Revenue Service, Notice 2025-67: qualified charitable distribution limit increased from $108,000 to $111,000 for 2026. irs.gov

Author disclosure

Educational content and personal opinion on public policy, not tax, legal, or investment advice. The calculator is a teaching model with a single elasticity parameter and is not a revenue estimate. Confirm the treatment of any gift with a tax professional.

More in Tax Strategy

Browse all tax strategy guides
Share

Get new guides by email

Evidence-based, no jargon. At most two emails a month. Unsubscribe any time.

Try it in Summitward

See multi-year tax projection in action with your own financial data. Free to start, no credit card required.

Disclaimer: This tool is for educational and informational purposes only and is not financial, investment, tax, or legal advice. Summitward is not a registered investment adviser, broker-dealer, or financial planner, and no fiduciary relationship is created by your use of it. Consult a qualified professional before acting. Past performance and model projections do not guarantee future results. Provided as is, without warranty of any kind; see our Terms of Service for limitations of liability.