This Congressional Budget Office (CBO) report was prepared at the request of the Ranking Member of the House Committee on Ways and Means. The report briefly reviews the history of the standard deduction in the federal income tax, examines patterns of charitable giving by itemizers and nonitemizers, and looks at the economic incentives that the tax system offers for charitable giving. It also discusses options for extending the deductibility of charitable donations to all taxpayers, focusing on revenue costs and the effects on levels of giving. In accordance with CBO's mandate to provide objective, nonpartisan analysis, this report makes no recommendations.

Robert McClelland of CBO's Tax Analysis Division prepared the report under the supervision of G. Thomas Woodward and Roberton Williams. Kurt Siebert provided assistance with data and figures. This paper benefited from comments by reviewers outside CBO, including Eugene Steuerle and Jane Gravelle. Within CBO, William Gainer, Arlene Holen, Deborah Lucas, John Sturrock, David Weiner, and Dennis Zimmerman provided comments.

Christine Bogusz edited the report, and Leah Mazade proofread it. Kathryn Winstead prepared the report for publication, Lenny Skutnik produced the printed copies, and Annette Kalicki prepared the electronic versions for CBO's Web site.

Dan L. Crippen
Director
December 2002

Summary and Introduction

Under current law, taxpayers can deduct from taxable income gifts they make to qualified charitable organizations only if they itemize their deductions on their tax return. Roughly 30 percent of taxpayers itemize deductions and can thus deduct charitable contributions; the other 70 percent receive no explicit tax benefit for their contributions. Both the Administration and some Members of Congress have proposed that taxpayers who elect the standard deduction ("nonitemizers") be allowed to deduct a limited amount of charitable contributions. Under the Administration's fiscal year 2003 budget proposal, for example, the annual deduction would be limited to $100 ($200 for joint filers) in 2002 through 2004, rising incrementally to $500 ($1,000 for joint filers) after 2011. The Administration asserts that the change "would help increase support for charitable organizations by rewarding and encouraging giving by all taxpayers."⁽¹⁾ Because the proposal would reduce the amount of taxes people pay (by allowing them to increase the amount they deduct), it would also reduce the revenue taken in by the Treasury. The Administration estimates that its proposal would lower federal revenue by $32.6 billion over the 2003-2012 period. Estimates made by the Joint Committee on Taxation (JCT) put the 10-year revenue loss at $29.3 billion.

Being able to deduct contributions from their taxable income offers itemizers an incentive to give to charity by reducing the net cost of that giving. Therefore, extending deductibility to people claiming the standard deduction ($4,550 for single taxpayers in 2001) should increase contributions. Nonitemizers would also receive explicit recognition for the contributions they made.

Some proponents of extending charitable deductions contend that the equity of the income tax should be judged not on levels of income but on levels of consumption. Currently, a relatively generous taxpayer with, say, $40,000 in income and $4,000 in contributions who takes the standard deduction pays the same income tax as a taxpayer with $40,000 in income and no contributions who takes the standard deduction, even though the first taxpayer has less money available for private consumption. Therefore, in this view, extending to nonitemizers the deduction for charitable giving improves the equity of the income tax.

In principle, taxpayers who claim the standard deduction already receive recognition for their charitable contributions under current tax law. But they typically select that deduction because it exceeds their deductible expenditures. If the Congress was to allow nonitemizers to deduct contributions in addition to taking the standard deduction, their total deduction would further expand beyond their actual itemizable expenditures.

Several factors suggest that allowing nonitemizers to deduct charitable contributions may have a relatively small impact on the overall level of giving. Because nonitemizers tend to face lower tax rates than itemizers do, they would tend to have a weaker tax incentive to contribute. Although estimates of the sensitivity to such incentives vary widely, taxpayers who take the standard deduction may well be relatively insensitive to those incentives. Thus, allowing them to deduct charitable contributions would be unlikely to lead to a very large increase in giving. Furthermore, even a large proportionate increase in giving by nonitemizers would be a small proportionate increase in total giving. The Congressional Budget Office (CBO) estimates that nonitemizers gave about 20 percent of all contributions in 1997, so a 20 percent increase in their giving would have increased overall giving by only 4 percent.

Two additional factors would raise the cost of proposals (in terms of revenue loss) without substantially increasing contributions. First, some taxpayers who itemize their deductions could reduce their taxes by switching to the standard deduction and claiming a charitable deduction, but their propensity to contribute would not change. Second, some taxpayers may claim unsubstantiated contributions, which happened in 1985 when a deduction for nonitemizers was allowed from 1982 through 1986.

This report looks at four ways of extending to nonitemizers the ability to deduct charitable contributions. The first option would allow them to claim an essentially unlimited deduction. That approach would result in the greatest increase in contributions but would also reduce revenue more than the other approaches would. Under a second option, contributions could be deducted up to a maximum amount, such as $100. That alternative would be less expensive than an unlimited deduction; but since it offered no incentive to contribute beyond the ceiling, it would probably stimulate contributions far less effectively than an unlimited deduction would. Also, verifying such small amounts of giving could prove difficult for the Internal Revenue Service (IRS) if large numbers of taxpayers claimed the deduction. The third and fourth options would allow taxpayers to deduct contributions only to the extent that they exceeded some threshold, such as $250 or a fixed percentage of their adjusted gross income (AGI), respectively. Allowing deductions above a floor amount would raise more contributions for any given revenue loss than deductions below a ceiling would. However, a floor would offer little encouragement to taxpayers who currently do not contribute or who could not afford to contribute more than the floor amount. All four options would be likely to increase overall contributions by less than 4 percent, and their primary effect, as is the case with most deductions, would be to reward taxpayers for their existing behavior.
 

History of the Charitable Deduction

Almost from its inception in 1913, the federal income tax has allowed taxpayers to subtract from their taxable income amounts spent for particular uses. For example, beginning in 1917, taxpayers could deduct donations made to charitable causes. To claim the deduction, taxpayers had to itemize their allowable expenditures. That itemization imposed a burden on taxpayers, but relatively few people were affected because only about 5 percent of households had to file tax returns.

World War II dramatically increased the reach of the income tax: by 1944, nearly three-fourths of households had to pay the tax. With that expansion came concern about the complexity of tax filing. To simplify tax returns, in 1944 the Congress created the standard deduction, then equal to 10 percent of a taxpayer's annual income, up to a maximum of $500. Taxpayers could select the standard deduction as an alternative to itemizing their expenditures on specific activities, reducing their taxes as if they had made that level of deductible expenditures but without having to comply with recordkeeping and reporting requirements. By taking the standard deduction, people are generally claiming deductions that are greater than their actual expenditures would have been if they had itemized.

A substantial majority of taxpayers claim the standard deduction today, thus simplifying both the preparation of tax returns and their auditing by the IRS. In 2000, for example, about 64 percent of tax returns had claims for the standard deduction.

As part of the Economic Recovery Tax Act of 1981, the Congress created a deduction for charitable contributions made by nonitemizers. The deduction was designed to be phased in over five years and then to expire. In 1982 and 1983, nonitemizers could deduct 25 percent of their first $100 of charitable contributions. In 1984, the limit was raised to $300. In 1985, 50 percent of all contributions were deductible; in 1986, nonitemizers could deduct 100 percent of all contributions. The deduction expired in 1987.

By 1984, however, the Treasury Department had concluded that such a deduction should not be extended because "the contribution deduction to non-itemizers creates unnecessary complexity, while probably stimulating little additional giving and presenting the IRS with a difficult enforcement problem."⁽²⁾

During the Congressional discussion surrounding the Tax Reform Act of 1986, the House attempted to address those issues by voting to keep a deduction for contributions above a floor of $100:

Allowing a non-itemizer deduction for relatively low levels of giving may impose recordkeeping burdens and complexity for many short-form filers, and create enforcement problems for the Internal Revenue Service. Accordingly, the committee has concluded that the non-itemizer deduction should be available only to the extent that the taxpayer's total charitable contributions for the year exceed $100.⁽³⁾

Ultimately, however, the Congress chose not to extend the deduction in the Tax Reform Act of 1986.

In the past several years, the Congress has expressed renewed interest in a charitable deduction for nonitemizers, crafting several proposals to that effect. For example, H.R. 1338 (from the 105th Congress) would have amended the tax code to allow nonitemizers to deduct total charitable contributions in excess of $1,000 ($2,000 for joint filers). The Administration's 2002 budget proposal included a nonitemizer deduction for total annual contributions less than or equal to the standard deduction; its 2003 proposal includes a deduction capped at $100 ($200 for joint filers) in 2002 through 2004, rising to $500 ($1,000) in 2012 (see Box 1). In the current Congress, H.R. 7, as approved by the House (the Community Solutions Act of 2001) would allow an annual deduction capped at $25 ($50 for joint filers) in 2002 and 2003, rising to $100 ($200 for joint filers) in 2010. As reported by the Senate Finance Committee, H.R.7 (the CARE Act of 2002) would allow a deduction for total contributions that exceeded $250 ($500 for joint filers) in 2002 and 2003. The deduction would be limited to $250 ($500 for joint filers) each year.

Issues in Allowing Nonitemizers to Deduct Charitable Contributions

Reimplementation of a deduction for nonitemizers raises several key questions.⁽⁴⁾ Which taxpayers make charitable contributions, and to whom do they give? To what extent will the deduction increase charitable giving? How will itemizers respond? Will the deduction complicate tax filing, and will tax evasion increase?

Who Contributes, and To Whom Do They Give?
CBO estimates that individuals contributed about $120 billion to charitable causes in 1997. Of that amount, two-thirds (about $78 billion) was contributed by taxpayers in the top quintile--or fifth--of the income distribution, based on adjusted gross income (see Figure 1). The fourth quintile contributed almost half of the remainder, or about $20 billion. The bottom three quintiles collectively contributed about $21 billion. Charitable contributions by nonitemizers rose slowly across income quintiles. In contrast, contributions by taxpayers who itemize their deductions rose dramatically with income, reflecting not only an increasing number of itemizers and higher average giving as income increased but also a response to the incentive to contribute provided by the deductibility of those contributions. Those factors resulted in itemizers in the top quintile accounting for about 60 percent of all charitable contributions made in that year.
 

Figure 1.

Charitable Contributions by Itemizers and Nonitemizers, by AGI Quintile, 1997

(Billions of dollars)

Source: Congressional Budget Office based on analyses of the 1998 Survey of Consumer Finances, the 1997 Consumer Expenditure Survey, and 1997 Statistics of Income data.

Notes: Total charitable contributions by itemizers were $99 billion in 1997; total contributions by nonitemizers equaled $21 billion in that year.

Each quintile represents one-fifth of the income distribution.

Some of the estimated $99 billion in charitable contributions made by itemizers in 1997 would have been made in the absence of any tax incentive. All of the $21 billion in contributions by taxpayers who claimed the standard deduction were made without any tax incentive.

Low-income taxpayers tend to take the standard deduction more than higher-income taxpayers do. More than 95 percent of taxpayers in the lowest income quintile claimed the standard deduction in 1997, whereas less than 20 percent of those in the top income quintile claimed it (see Figure 2). About 80 percent of taxpayers who claimed the standard deduction in that year were in the lowest three quintiles.
 

Figure 2.

Percentage of Taxpayers Claiming the Standard Deduction, by AGI Quintile, 1997

(Percent)

Source: Congressional Budget Office analysis of 1997 Statistics of Income data.

Note: Each quintile represents one-fifth of the income distribution.

Taxpayers who claim the standard deduction tend to give to the same types of organizations as taxpayers who itemize, according to survey data from 1996 (see Figure 3). For example, both types of taxpayers give approximately 60 percent of their contributions to religious organizations. Contributions to human services organizations, health institutions, and educational institutions constitute roughly one-quarter of the total for both itemizers and nonitemizers.
 

Figure 3.

Recipients of Charitable Contributions by Nonitemizers and Itemizers, 1995

Source: Congressional Budget Office based on data from Giving and Volunteering in the United States (1996).

Those numbers should be interpreted with caution, however. The same data indicate that the percentage of contributions going to religious organizations is larger for households with lower income. If nonitemizers tend to have lower income than itemizers have, that pattern suggests that nonitemizers should tend to give a greater proportion of their contributions to religious organizations.

Does the Deduction Stimulate Contributions?
The ability to deduct charitable contributions offers itemizing taxpayers an incentive to contribute, regardless of their income, by reducing the cost of giving (net of tax savings). For example, consider a taxpayer in the 15 percent tax bracket who itemizes her deductions. If she contributes an additional dollar, in most cases she may increase her deductions by a dollar.⁽⁵⁾ That increase lowers her taxable income by one dollar and her taxes by 15 cents, so the after-tax cost of her additional donation of one dollar (the "tax price") is only 85 cents. An otherwise identical taxpayer who claims the standard deduction cannot reduce her taxable income, so her cost of donating an additional dollar is one dollar.

Itemizing taxpayers in higher tax brackets face a stronger incentive to give because their tax price is lower: in 1999, it was 72 cents for a taxpayer in the 28 percent bracket, 69 cents for a taxpayer in the 31 percent bracket, and so on.⁽⁶⁾ The progressivity of the tax code (rates increase as income rises), along with the income distribution of taxpayers taking the standard deduction, implies that nonitemizers tend to face lower tax rates; therefore, they would have a relatively weak incentive to give if they were allowed to deduct their charitable contributions. In 1999, about 70 percent of nonitemizers faced a 15 percent tax rate and hence would see their tax price of giving decrease by 15 percent, from one dollar to 85 cents (see Table 1). Less than 1 percent of itemizers faced a rate above 28 percent. In contrast, only about 40 percent of itemizers were in the 15 percent tax bracket, and about the same percentage were in the 28 percent bracket. The average tax rate for nonitemizers was almost 16 percent, whereas taxpayers who itemize had an average rate of about 23 percent.

Table 1.
Percentage of Taxpayers in Each Tax Bracket and the Associated Statutory Tax Price of Giving, 1999
Tax Bracket	Statutory Tax Price (Dollars)		Nonitemizers (Percent)		Itemizers (Percent)
No Taxable Income	1.00		17.3		6.3
15 Percent	0.85		69.9		39.3
28 Percent	0.72		11.9		40.4
31 Percent	0.69		0.6		7.0
36 Percent	0.64		0.1		3.0
39.6 Percent	0.604		0.1		1.9
Alternative Minimum Tax	0.76/0.72		0.1		2.2
Memorandum:
Average Tax Ratea	n.a.		15.8		23.0
Source: Congressional Budget Office analysis of 1999 Statistics of Income data.
Note: n.a. = not applicable.
a. The average tax rate uses statutory marginal rates modified by tax provisions limiting deductions and personal exemptions.

Research done before 1995 suggested that contributions by itemizing taxpayers were sensitive to tax rates. That sensitivity--measured as the percentage change in contributions from a 1 percent change in the tax price--is known as a price elasticity. Estimated elasticities ranged from -1.0 to -1.4, with many near -1.2.⁽⁷⁾ At that time, it was believed that a 10 percent decrease in the tax price would increase contributions by about 12 percent. Those estimates generally applied to itemizing taxpayers, who tend to have high income.

The sensitivity of low-income individuals to tax rates was and is subject to much greater uncertainty.⁽⁸⁾

Reflecting the use of more-sophisticated statistical analyses, recent research suggests that the price elasticity of contributions by itemizers is more likely to be in the range of -0.4 to -0.8.⁽⁹⁾ If those estimates more accurately reflect taxpayer behavior, then a 10 percent decrease in the tax price would increase itemizers' contributions by between 4 percent and 8 percent.

The effect of the tax incentive on nonitemizers is much more difficult to estimate because all taxpayers who claim the standard deduction face the same tax price. Only by examining data from 1982 to 1986--when nonitemizers could deduct part or all of their charitable contributions--can researchers observe the effect of tax prices on nonitemizers. Because the maximum allowable deduction was quite low from 1982 to 1984, researchers are generally restricted to using data from 1985 (when one-half of all contributions were deductible by nonitemizers) and 1986 (when all contributions were deductible). Consequently, research on the sensitivity of nonitemizers to tax incentives reveals a wide range of estimates.

For example, a recent study estimated the elasticity of nonitemizers in 1985 and 1986 as -3.4.⁽¹⁰⁾ That number, however, is highly suspect, for several reasons. First, an enormous change in tax law occurred in 1986 with enactment of the Tax Reform Act. Taxpayers' behavior around that time would have reflected not only tax incentives for nonitemizers but also anticipated changes under the law. Second, when the study's authors (Dunbar and Phillips) used a different statistical method, they found no evidence that contributors would respond to tax incentives at all. Third, another researcher, using the same data, estimated the elasticity as -0.6 to -0.8.⁽¹¹⁾ And a recent Congressional Research Service report estimated an elasticity of only -0.1.⁽¹²⁾ Confronted by such a degree of uncertainty, researchers at the Urban Institute used a range of possible elasticities when estimating the effect of a deduction for nonitemizers.⁽¹³⁾ This paper follows that approach, using elasticities of -0.2, -0.6, and -1.0.

How Will Itemizers Respond?
If there was a charitable deduction for nonitemizers, some itemizing taxpayers would reduce their taxes by switching to the standard deduction, even though charitable contributions would increase little, if at all.⁽¹⁴⁾ For example, imagine a single taxpayer, with $800 in charitable contributions and $4,000 in other deductions, whose income is in the 27 percent bracket. Because the $4,800 in deductions is higher than the standard deduction of $4,550, his taxable income (and therefore his taxes) will be lower if he itemizes. His tax price of contributing an extra dollar is then 73 cents. But if the Congress enacted an unlimited deduction for nonitemizers, his tax bill would be lower if he claimed the standard deduction of $4,550 and a charitable deduction of $800 because his deductions would total $5,350. His taxes would be reduced by $148.50, and the tax price he faced would still be 73 cents. In other words, he would face no additional incentive to contribute but would decrease his taxes by $148.50. Note, however, that he would continue to itemize his deductions if the nonitemizer deduction was limited to a maximum of $200. In that case, the sum of the standard deduction ($4,550) and the $200 allowable deduction would be $4,750, less than the $4,800 in itemized deductions he could claim. Therefore, limiting the deduction for nonitemizers reduces the number of itemizing taxpayers who would benefit from switching to the standard deduction.

Nonetheless, even if they could deduct only small amounts of charitable contributions, a large number of itemizing taxpayers would have an incentive to switch to the standard deduction (see Figure 4). For example, if a maximum of $100 in contributions ($200 for joint filers) could have been claimed in 1997, about 450,000 itemizing taxpayers could have benefited from claiming the standard deduction and taking the nonitemizer deduction for charitable contributions. A $500 maximum ($1,000 for joint filers) could have induced about 1.3 million itemizing taxpayers to switch their status, and almost 2 million itemizing taxpayers could have benefited from switching to the standard deduction if the ceiling had been $1,000 ($2,000 for joint filers). Of course, not all of those itemizing taxpayers would necessarily have switched to the standard deduction immediately after the charitable deduction was allowed. However, it is reasonable to expect that over time, nearly all itemizers who could save money by switching to the standard deduction and then deducting charitable contributions would do so.
 

Figure 4.

Number of Itemizing Taxpayers in 1997 Switching to the Standard Deduction, by Maximum Allowable Charitable Deduction

(Millions of taxpayers)

Source: Congressional Budget Office analysis of 1997 Statistics of Income data.

Notes: This figure shows the number of itemizing taxpayers who would have benefited by switching to the standard deduction in 1997 (without increasing their charitable contributions) if a charitable deduction for nonitemizers had been allowed in that year.

The number of itemizers switching their status was calculated by using a maximum deduction for joint filers equal to twice that shown on the horizontal axis. For example, the number of switchers from a maximum deduction of $100 was calculated by assuming that taxpayers filing jointly had a $200 maximum deduction.

Consistent with that pattern, the potential revenue loss from itemizers switching to the standard deduction would increase as the limit on deductible contributions rose (see Figure 5). A charitable deduction of up to $100 ($200 for joint filers) in 1997 would have cost more than $50 million in lost revenue if all itemizers for whom not itemizing made sense actually switched. That loss would have increased to almost $140 million if the deduction limit had been $500 ($1,000 for joint filers) and nearly $300 million if the ceiling had been $1,000 ($2,000 for joint filers).
 

Figure 5.

Revenue Loss in 1997 from Itemizing Taxpayers Switching to the Standard Deduction, by Maximum Allowable Charitable Deduction

(Millions of dollars)

Source: Congressional Budget Office analysis of 1997 Statistics of Income data.

Notes: This figure shows the revenue loss that would have resulted in 1997 from itemizing taxpayers switching to the standard deduction if they would have benefited by doing so (without increasing their charitable contributions), assuming that a charitable deduction for nonitemizers had been allowed in that year.

The revenue loss was calculated by using a maximum deduction for joint filers equal to twice that shown in the horizontal axis. For example, the revenue loss from a maximum deduction of $100 was calculated by assuming that taxpayers filing jointly had a $200 maximum deduction.

If the maximum allowable deduction had equaled the standard deduction, the revenue loss could have been as large as $1.0 billion.

Will Tax Filing Be More Complicated, and Will Tax Evasion Increase?
Adding a deduction for charitable contributions for taxpayers claiming the standard deduction would increase the recordkeeping burden for some taxpayers but could reduce it for others. Contributors taking the standard deduction would have to keep records verifying their charitable contributions, whereas previously they did not. But the burden might be reduced for itemizers under current law who would be certain to take the standard deduction if contributions were deductible for nonitemizers; those taxpayers would no longer be required to keep records of all their other deductions.

When a charitable deduction was available to nonitemizers during the early to mid-1980s, the percentage of unverified contributions by nonitemizers was nearly twice the percentage by itemizers. Under the Tax Compliance Measurement Program (TCMP), a small number of randomly selected returns were carefully audited, and unsubstantiated deductions were disallowed.⁽¹⁵⁾ Because 50 percent of all contributions by itemizers and nonitemizers were deductible in 1985, the TCMP in that year may be used to compare the overreporting of the two groups.

Overall, less than 10 percent of contributions by itemizers were disallowed, compared with almost 18 percent for nonitemizers (see Table 2).⁽¹⁶⁾ Among nonitemizers, the difference between those with AGI below $25,000 and those with AGI above $25,000 is striking: 20.8 percent versus 8.2 percent, respectively. That difference may reflect the view of those taking the deduction that they faced little chance of being caught. As the Treasury Department wrote in 1984:

Non-itemizers generally make smaller charitable gifts than itemizers. A deduction may be claimed for numerous small gifts, made to a number of different organizations. It is extremely difficult and expensive for the Internal Revenue Service to monitor these deductions. Further, the cost of administration is disproportionate to the amount involved. These factors may prompt dishonest taxpayers to conclude that they can misrepresent their charitable gifts with impunity.⁽¹⁷⁾

An Analysis of Four Options

Four methods for extending the charitable deduction to nonitemizers have been proposed:⁽¹⁸⁾ allow taxpayers who claim the standard deduction to deduct all charitable contributions, allow them to deduct all contributions up to a maximum amount (or ceiling), allow them to deduct only contributions above a minimum dollar amount (or floor), or allow them to deduct only contributions above a minimum percentage of AGI. This report analyzes each approach using the data and methods described in Appendix A. It estimates the increased contributions per dollar of revenue loss (efficiency) that might result and examines in detail specific proposals for each option. Because of the uncertainty about how nonitemizers would adjust their level of charitable contributions in response to a shift in their tax price, the estimates use a range of elasticities.⁽¹⁹⁾ None of the estimates of revenue loss include the effects from tax evasion, so those estimates should be treated as lower bounds.

Of the four options, an essentially unlimited deduction would increase contributions the most but would be the costliest in terms of revenue loss. Lowering the cost by setting either a dollar or a percentage floor on deductions would increase contributions by far more than capping the deduction would (with the same revenue cost). In each case, most of the cost would come from itemizers switching to the standard deduction and from the new deductibility of contributions that would have been made even without the deduction. Revenue loss might be further mitigated either by lowering the existing standard deduction or by applying the same floor or ceiling on deductibility to the contributions of taxpayers who itemize their deductions. Before explaining each option in detail, this paper first discusses the options on the basis of their efficiency.

An Evaluation of Efficiency
A stated purpose of allowing nonitemizers to deduct contributions is to increase the overall amount of charitable giving. But that gain comes at the cost of forgone revenue. One measure of the efficiency of each option is the increase in giving it induces (measured as a percentage of the total revenue loss). Ideally, the increase in contributions would equal or exceed the loss of revenue. In reality, however, such a balance is unlikely: CBO estimates that an unlimited deduction would induce contributions equal to the revenue loss only if the tax price elasticity of nonitemizers was about -1.7, which is substantially greater in absolute value than most elasticity estimates.⁽²⁰⁾ Although increased contributions would probably be less than the revenue loss they induced, some methods for implementing a deduction for nonitemizers are more efficient than others.

For example, an unlimited deduction would undoubtedly increase contributions more than any other option would. If an unlimited deduction had been in place in 1997, it would have raised contributions by as much as $3.5 billion, CBO estimates, depending on how responsive nonitemizers were to a change in the tax price of giving (see Table 3). That increase would have been accompanied by the greatest reduction in revenue of any option, as much as $5.6 billion in 1997. If nonitemizers had had a tax price elasticity of -0.6 (instead of -1.0), contributions would have increased by about 40 percent of the revenue loss to the Treasury (see Figure 6).⁽²¹⁾

Table 3.
Effects in 1997 of Allowing a Nonitemizer Deduction for Contributions Less Than or Equal To the Standard Deduction
									Change in Contributions
	Sources of Revenue Loss (Billions of dollars)								Increased Contribu- tions (Billions of dollars)		Increase in Total Contribu- tions (Percent)
Elasticity	Increased Giving		Existing Giving		Itemizers Switching Status		Total Loss
0	0		3.8		1.0		4.9		0		0
-0.2	0.1		3.8		1.0		5.0		0.7		0.6
-0.6	0.4		3.8		1.0		5.3		2.1		1.8
-1.0	0.7		3.8		1.0		5.6		3.5		2.9
Source: Congressional Budget Office.
Note: For an explanation of the final row of this table, see Appendix B.

 

Figure 6.

Estimated Efficiency of Deduction Floors and Ceilings Assuming an Elasticity of -0.6, by Total Revenue Loss, 1997

(New contributions per dollar of revenue loss, in percent)

Source: Congressional Budget Office.

Note: The curves in this figure were created by interpolating among a small number of data points. Because the sampling error may be substantial and varies from point to point, those curves should be viewed as rough approximations.

Capping the allowable deduction or allowing deductions only for donations above a floor amount would reduce the revenue loss but would also yield a smaller amount of new contributions. While both methods would reduce revenue and contributions, their relative efficiency would differ.

Consider first a ceiling on deductions. If the ceiling equaled the standard deduction, it would be equivalent to an unlimited deduction for nonitemizers. That is because under current law, any taxpayer giving more than the standard deduction would have been better off itemizing deductions. Although a lower ceiling would reduce the amount that taxpayers making large contributions were able to deduct, it would still provide those taxpayers with some tax relief. It would not, however, provide them with an incentive to increase their contributions. Progressively lower ceilings would provide a tax incentive to a progressively smaller group of taxpayers and offer tax cuts without that incentive to a progressively larger group of taxpayers. For any given level of responsiveness to the charitable deduction, the amount of new giving would decline faster than the revenue loss, and the option's efficiency--the new giving as a percentage of revenue loss--would fall (see the lower line on Figure 6). In 1997, cutting the revenue loss in half by lowering the ceiling to $900 ($1,800 for joint filers) might have reduced the efficiency of the option from 40 percent to less than 30 percent.

In contrast, imposing a floor on deductible contributions could lead to increased efficiency as revenue cost declined. If the floor was set at zero, the option would be equivalent to an unlimited deduction in terms of both revenue loss and generation of new donations. A floor above zero would exclude taxpayers with small contributions, eliminating the tax incentive and reducing the revenue loss. Excluding small contributors would slightly increase efficiency if their tax rates--and hence their incentive to give--tended to be low.⁽²²⁾ Taxpayers making contributions above the floor amount would still be encouraged to give, but they would only receive a tax break on contributions exceeding the floor, again increasing efficiency. Progressively higher floors would provide a tax incentive to a progressively smaller group of taxpayers, reducing the revenue loss and slightly raising efficiency (see the top line on Figure 6). In 1997, cutting the revenue loss in half by raising the floor to $1,000 ($2,000 for joint filers) might have modestly raised the efficiency of the option, from 40 percent to almost 45 percent.

The Four Options
This paper looks at four specific proposals to extend the deductibility of charitable contributions to nonitemizers. They are allowing an unlimited deduction, capping the deduction at $100, setting a $250 floor on deductions, or setting a floor on deductions of 2 percent of adjusted gross income.

Allow an Unlimited Deduction. An unlimited deduction in 1997 for nonitemizers would have increased total charitable contributions by $0.7 billion to $3.5 billion, CBO estimates, an increase of between 0.6 percent and 2.9 percent. (As mentioned previously, unlimited refers to contributions less than or equal to the standard deduction.) The Treasury Department and the Joint Committee on Taxation have estimated the 10-year cost of a phased-in version of such a deduction proposed by the Administration for 2002 as $52 billion and $84 billion, respectively (see Box 1). Had that deduction been fully phased in in 1997, CBO estimates, tax revenue would have dropped by between $5.0 billion and $5.6 billion (see Table 3).⁽²³⁾ Of that loss, $100 million to $700 million would have resulted from new contributions. The first row in Table 3 shows the revenue loss that would have occurred even if contributions did not change at all. That revenue loss would have come from subsidizing existing donations ($3.8 billion) and from taxpayers switching from itemizing deductions to taking the standard deduction ($1.0 billion).

Cap the Deduction at $100. Limiting the amount of contributions that a nonitemizing taxpayer could deduct would reduce the loss of revenue. For example, H.R. 7 would limit the deduction to $25 ($50 for married taxpayers filing a joint return) in 2003 and would increase the maximum to $100 ($200 for joint filers) in 2010. The low ceiling on deductions would dramatically reduce the revenue loss from both of those proposals: JCT estimates that H.R. 7 would lower revenue by $1.3 billion when fully phased in in 2010 (see Table 4). If a limit of $100 ($200 for joint filers) had been in effect in 1997, the revenue loss in that year, assuming no overstatement of contributions, would have been $500 million or less (see Table 5). The losses from taxpayers deducting existing contributions and from itemizers switching to the standard deduction would have been about $400 million. The Administration's 2003 budget proposes a phased-in ceiling that would reach $500 in 2012 ($1,000 for joint filers), which would reduce revenue by $29.3 billion to $32.6 billion over 10 years (see Box 1).

Table 4.
JCT's Estimates of the Revenue Effects of the Charitable Giving Incentive Provisions of H.R. 7
	2002	2003	2004	2005	2006	2007	2008	2009	2010	2011	Total, 2002- 2006	Total, 2002- 2011
Estimated Revenue Effect	-40	-269	-316	-561	-573	-625	-876	-894	-966	-1,252	-1,759	-6,372
Source: Joint Committee on Taxation, "Estimated Revenue Effects of a Chairman's Amendment in the Nature of a Substitute to H.R. 7, the 'Community Solutions Act of 2001,' Scheduled for Markup by the Committee on Ways and Means on July 11, 2001" (July 2001).

Table 5.
Effects in 1997 of Allowing a Nonitemizer Deduction for Contributions Less Than or Equal To $100 ($200 for Joint Filers)
									Change in Contributions
	Sources of Revenue Loss (Billions of dollars)								Increased Contribu- tions (Billions of dollars)		Increase in Total Contribu- tions (Percent)
Elasticity	Increased Giving		Existing Giving		Itemizers Switching Status		Total Loss
0	0		0.4		0.1		0.4		0		0
-0.2	*		0.4		0.1		0.4		*		**
-0.6	*		0.4		0.1		0.4		*		**
-1.0	*		0.4		0.1		0.5		*		**
Source: Congressional Budget Office.
Note: * = less than $50 million; ** = less than 0.05 percent.

Although the maximum deduction would be capped under those proposals, the incentive to contribute would still apply to contributions under the cap. In that way, taxpayers who currently do not contribute to charitable causes would be encouraged to start giving. Those taxpayers tend to have low income, so the deduction could induce many low-income taxpayers to contribute. However, for the incentive to apply, they must owe federal income taxes. Also, because low-income taxpayers tend to face low tax rates, their incentive to contribute would be weaker than the incentive faced by taxpayers with higher income.

This option would result in little additional giving by current contributors because it would not give them an incentive to make contributions above the ceiling. CBO estimates that if a ceiling of $100 ($200 for joint filers) had been in place in 1997, it would probably have spurred total additional contributions of less than $50 million, or one-twentieth of 1 percent of total contributions, because the majority of contributors were already giving at least $100 or $200 per year (see Table 5). A recent study estimated that about 70 percent of nonitemizers who contributed in 1999 gave more than $100, and about 45 percent contributed more that $200.⁽²⁴⁾ Although those taxpayers would have had more after-tax income if they had itemized their deductions, their incentive to contribute would have remained the same. In addition, contributions would have been subject to recordkeeping requirements. It should also be noted that a small increase in actual giving may be accompanied by a large increase in claimed contributions if taxpayers believe that reporting the maximum allowable contribution is both routine and unquestioned by the IRS.

Set a $250 Floor on Deductions. Allowing deductions only for contributions above a floor amount would reduce the problems (noted above) caused by taxpayers reporting small donations. A floor would result in less revenue loss than an unlimited deduction would, while still providing many contributors with an incentive to give. A deduction above a floor of $250 ($500 for joint filers) in 1997 would have reduced revenue by between $4.6 billion and $5.1 billion, CBO estimates (see Table 6). Revenue loss would be substantially lower under this option than it would be under an unlimited deduction; itemizers switching to the standard deduction would reduce revenue by only $0.8 billion, and the deduction for existing contributions by nonitemizers would lower revenue by $3.6 billion.

Table 6.
Effects in 1997 of Allowing a Nonitemizer Deduction for Contributions in Excess of $250 ($500 for Joint Filers)
									Change in Contributions
	Sources of Revenue Loss (Billions of dollars)								Increased Contribu- tions (Billions of dollars)		Increase in Total Contribu- tions (Percent)
Elasticity	Increased Giving		Existing Giving		Itemizers Switching Status		Total Loss
0	0		3.6		0.8		4.4		0		0
-0.2	0.1		3.6		0.8		4.6		0.7		0.5
-0.6	0.4		3.6		0.8		4.8		2.0		1.7
-1.0	0.7		3.6		0.8		5.1		3.3		2.8
Source: Congressional Budget Office.

Because taxpayers who make small or no charitable contributions would have no tax incentive to give, this option would not increase contributions as much as an unlimited deduction would. However, because small contributions account for only a minor part of the lost revenue, the difference between the two options is relatively modest (a range of $0.7 billion to $3.3 billion for the $250 floor option versus a range of $0.7 billion to $3.5 billion for the unlimited deduction option). This option also would require fewer taxpayers to keep records of contributions and would reduce the opportunity for dishonest taxpayers to claim small or nonexistent contributions.⁽²⁵⁾

Set a Floor on Deductions of Two Percent of Adjusted Gross Income. Alternatively, the incentive to contribute could be extended to lower-income taxpayers by defining the floor as a percentage of AGI, such as 2 percent, instead of a fixed dollar amount. Taxpayers not contributing would still not face an incentive to contribute small amounts, but very low-income taxpayers would find it easier to reach the minimum necessary for the incentive to start. For married taxpayers filing jointly with annual income of less than $25,000, the floor would be lower than $500, whereas the floor would be relatively higher for households with income above $25,000. According to JCT, the estimated cost for a deduction with a floor equaling 2 percent of AGI (but not exceeding the standard deduction) over the 2003-2012 period would be $19.3 billion. Had this deduction been in effect in 1997, the cost would have been between $3.7 billion and $4.2 billion in that year, CBO estimates (see Table 7).⁽²⁶⁾ The revenue loss from taxpayers deducting existing contributions and from itemizers changing to the standard deduction would be even lower with a percentage floor than with a dollar floor; the deduction of existing contributions would reduce revenue by $3.0 billion, and itemizers switching their status would lower revenue by $0.5 billion.

Table 7.
Effects in 1997 of Allowing a Nonitemizer Deduction for Contributions in Excess of Two Percent of Adjusted Gross Income
									Change in Contributions
	Sources of Revenue Loss (Billions of dollars)								Increased Contribu- tions (Billions of dollars)		Increase in Total Contribu- tions (Percent)
Elasticity	Increased Giving		Existing Giving		Itemizers Switching Status		Total Loss
0	0		3.0		0.5		3.6		0		0
-0.2	0.1		3.0		0.5		3.7		0.6		0.5
-0.6	0.4		3.0		0.5		3.9		1.7		1.4
-1.0	0.6		3.0		0.5		4.2		2.9		2.4
Source: Congressional Budget Office.

A percentage floor has several similarities to a dollar floor. As with a dollar floor, a percentage floor would encourage more contributions than a capped deduction would. In 1997, contributions under such an option would have grown by $0.6 billion to $2.9 billion, compared with less than $50 million for a ceiling of $100 ($200 for joint filers). Also, like a dollar floor, a percentage floor would impose a recordkeeping burden on fewer taxpayers than an unlimited deduction would. Tax evasion might be less of an issue, too, because a floor would eliminate small deductions, which are harder for the IRS to verify.

Additional Ways to Reduce Revenue Loss
Allowing taxpayers to deduct only contributions below a maximum or above a minimum would reduce the loss in tax revenue without altering the standard deduction or itemized deductions that they might claim. Revenue loss could be further reduced by changing either of those kinds of deductions.

Lowering the standard deduction could be one way to mitigate revenue loss. For example, if a maximum of $100 was deductible for nonitemizers, the standard deduction could be reduced by $100. That change would keep the same incentives and same recognition for existing charitable contributions, but it could eliminate the loss in revenue and actually raise revenue if some people claimed less than the maximum deduction. On the downside, nonitemizers who gave less than the cap would lose some of their current deductions. Alternatively, the standard deduction could be reduced by the estimated average level of contributions that a given option would induce.

The cost of lost revenue could also be reduced by extending limits on deductibility to itemizers. For example, JCT has estimated that limiting the charitable deduction for itemizers to giving in excess of 2 percent of AGI would raise more than $130 billion between 2002 and 2011.⁽²⁷⁾ That figure and figures from the previous section suggest that a floor of 2 percent of AGI applied to the contributions of all taxpayers could potentially raise revenue while extending the deduction to nonitemizers.

The Congressional Budget Office (CBO) used three separate data sets for the analysis in this report. Data on charitable contributions by taxpayers who currently itemize their deductions come from the 1997 Individual Statistics of Income (SOI). Data on contributions in excess of $500 made by taxpayers claiming the standard deduction are from the 1998 Survey of Consumer Finances (SCF). Contributions under $500 were imputed to the SCF using data from the 1997 Consumer Expenditure Survey (CE).

Estimates of the number of itemizing taxpayers who would switch to the standard deduction if a charitable deduction for nonitemizers was allowed used nondependent returns from the 1997 SOI data. Taxes and tax rates were estimated both with and without the applicable deduction for nonitemizers. If taxes were lower using the standard deduction, CBO assumed that taxpayers would switch to that deduction; otherwise, continued use of the itemized deduction was assumed. The calculations of taxes did not include phase-outs of deductions and personal exemptions, the alternative minimum tax, or tax credits, such as the earned income tax credit.

CBO created nonitemizer returns using each respondent's declared itemization status from the 1998 SCF data. Information on contributions was limited to cash contributions in excess of $500 per year. Smaller contributions were imputed from the 1997 CE, and taxes were calculated using the National Bureau of Economic Research's TAXSIM program.

The 1997 CE consists of five quarterly interviews conducted throughout 1997 and the first quarter of 1998. Total contributions were calculated as the sum of contributions to religious organizations, educational institutions, and "charities, such as United Way, Red Cross, etc."⁽²⁸⁾ Itemization status was determined using TAXSIM. Contributions below $500 per year were imputed from the CE to the SCF by first dividing both data sets into income quintiles, or fifths. For a given record in the SCF with contributions of less than $500, a record from the CE in the same income quintile with contributions under $500 was randomly chosen with replacement, with the probability of selecting a given CE record proportional to the survey weight of that record. Because each primary economic unit in the SCF was represented by five separate records, in effect five imputations were performed for each unit in the SCF with annual contributions below $500. Taxes on the SCF with imputed contributions also were calculated using TAXSIM.

Once contributions and tax revenue from each individual record were known, the loss from a nonitemizer deduction was calculated. Increased contributions were estimated by applying the assumed elasticities to existing contributions and applicable tax rates.

CBO estimated losses from existing contributions by using the minimum of contributions and the applicable ceiling. The percentage change in price was calculated as the marginal tax rate on the last dollar of giving so that a taxpayer in the 15 percent bracket who claimed the standard deduction would face a 15 percent decrease in his or her tax price. Increased contributions from a deduction with a floor were calculated in the same manner as the other options, except that the increase was mitigated by an income effect. That income effect, which adjusted for the extra taxes that must be paid because contributions are not fully deductible, was estimated as 2 percent of the additional taxes paid because of the floor.

Readers familiar with the idea that a unit price elasticity implies that the increase in contributions equals the loss in revenue (excluding the loss from itemizers who switch to the standard deduction) may be confused by the final row of numbers in Table 3, in which the $3.5 billion increase in contributions is closer to the $3.8 billion revenue loss from a tax break for existing giving than to the loss from increased giving and existing giving combined. The two numbers may be reconciled by recognizing that the standard elasticity proof only holds for infinitesimal changes in the tax rate. For discrete changes, the standard proof requires that the percentage change in taxes be calculated from the ending tax price, not the initial tax price.

That calculation may be shown as follows. First, assume that contributions C are not deductible and are subject to a single tax rate t. Without deductibility, revenues from this amount are then R = t × C, and the tax price of contributing an additional dollar is 1. Assume now that contributions become deductible. Then the revenue loss is
 

dR = dt × C + t × dC

(1)

where dt is the change in the applicable tax rate. For a taxpayer in the 15 percent tax bracket, this change would be 0.15 - 0 = 0.15. To see that a unit price elasticity exists when the revenue loss equals the increase in contributions, substitute dC for dR in equation (1) and rearrange it so that
 

[(1 - t)/C]dC/dt = 1.

(2)

Note that the tax price p is now (1 - t) and that dC/dt = -dC/dp so that
 

[p/C]dC/dp = -1,

(3)

or the tax price elasticity is unity. Note that the implicit percentage change in price dp/p is being calculated at the point (1 - t). For infinitesimal changes, such as when t = 0.01, 1 and 1 - t are approximately the same number, and a unitary elasticity implies that increased contributions offset revenue loss. For discrete changes, the proof no longer holds. For example, a taxpayer in the 15 percent bracket experiences a price change from 1 to 0.85, or a percentage change of (1 - 0.85)/1, not (1 - 0.85)/0.85. For the discrete change considered here, the appropriate elasticity measure is [1/C]dC/dp. If revenue neutrality holds, the implied elasticity is therefore
 

[1/C]dC/dp = -1/p,

(4)

or about -1.18 for a taxpayer in the 15 percent bracket. Thus, for such a taxpayer the elasticity at which increased contributions equal the revenue loss (excluding switchers) is actually -1.18, not -1. For a taxpayer in the 28 percent bracket, the necessary elasticity is about -1.39; for a taxpayer in the 32 percent bracket, it is -1.47.

It follows that if the price elasticity is unity when calculated at a price equal to 1, then the revenue loss will exceed the increase in contributions. The difference between the two may be found by setting the elasticity to unity
 

[1/C]dC/dp = -1

(5)

and then solving for C and substituting into (1)
 

dR = dC(1 + t).

(6)

For a taxpayer in the 15 percent bracket with a unitary elasticity as calculated in equation (5), the revenue loss will exceed the increase in contributions by about 15 percent.