The Impact of Five Different Tax Policy Changes on Household Giving in the United States

Abstract About $450 billion were donated to U.S. nonprofits in 2019 according to the most recently available data (Giving USA Foundation 2020). However, despite the increases in charitable dollars, the share of households that donate has been declining: in 2000, 67 percent of American households donated to nonprofits, but in 2016, only 53 percent of American households donated (Indiana University Lilly Family School of Philanthropy 2019). This trend in decreasing share of U.S. households that donate to charitable causes pre-dates the passage of the 2017 Tax Cuts and Jobs Act (TCJA), but could be accelerated by the recent policy changes. TCJA significantly changed federal tax policy and these changes are expected to affect charitable giving (Brill and Choe 2018; Ricco 2018; Rooney et al. 2017). Nonprofit leaders, as well as policymakers, have been exploring additional policy proposals to offset the potential negative impact on charitable giving. This paper investigates the estimated effects of potential policy proposals on charitable giving, donor incidence rates, and Treasury revenue. This study used the Penn Wharton Budget Model (Penn 2019a, 2019b) to run microsimulations of the effects of five tax policy proposals on charitable giving dollars, the number of households that donate, and the forgone Treasury revenue. The five proposals included: a non-itemizer charitable deduction; a non-itemizer charitable deduction with a cap; a non-itemizer charitable deduction with a floor; an enhanced non-itemizer charitable deduction, which provides a higher value deduction for low- and middle-income households; and a non-itemizer non-refundable 25 percent charitable giving tax credit. Of the five policy options analyzed, providing a non-refundable 25% charitable giving tax credit to non-itemizers has the largest positive impact, increasing both the amount of charitable giving dollars ($37 billion in 2018 dollars) and the number of donor households (10.6 million) of the five policy options analyzed. However, it is also the most “expensive” proposal (measured in terms of forgone Treasury revenue) for United States (U.S.) Treasury revenue (−$33.0 billion). Four of the five policy proposals bring in more charitable dollars than are lost in Treasury revenue. Four of the five policy proposals bring in more charitable dollars than were projected to have been lost as a result of TCJA. All five proposals bring in more donor households that were expected to be lost as a result of TCJA. This paper is based on a published report written and researched by [school] in partnership with the Wharton School of Business at the University of Pennsylvania and commissioned by Independent Sector. The report, “Charitable Giving and Tax Incentives Estimating changes in charitable dollars and number of donors resulting from five policy proposals,” can be found at this link: http://hdl.handle.net/1805/19515.


Introduction and Statement of the Issue
In 2019, charitable giving by individuals, corporations, and foundations in the United States reached $449.64 billion, 1 which included $309.66 billion from individuals (Giving USA Foundation 2020). Overall, charitable giving dollars and giving by individuals have generally been increasing over time, with some decreases in recession years (Giving USA Foundation 2020). However, the percentage of American households who donate has been declining. In 2002, 68.5 percent of Americans donated to nonprofits, 2 while in 2016, only 53.1 percent of Americans donated (Clark, Han, and Osili 2019; Indiana University Lilly Family School of Philanthropy 2019). Therefore, charitable giving appears to be increasingly concentrated among high-income households, who already dominate the charitable giving landscape (Osili, Clark, and Bergdoll 2018;Osili and Zarins 2018; Indiana University Lilly Family School of Philanthropy 2019; Reich 2006Reich , 2018Rooney 2018Rooney , 2019. This has led to concerns among nonprofit leaders, policymakers, and academics about how to increase equity while addressing efficiency (Giridharadas 2018;Reich 2018). This paper investigates the estimated effects of potential policy proposals compared to current 2018 tax law on charitable giving, donor incidence rates, and Treasury revenue.
We used the Penn Wharton Budget Model (Penn 2019a(Penn , 2019b to run microsimulations of the effects of five tax policy proposals on charitable giving dollars, the number of households that donate, and Treasury revenue. Results presented are estimates for 2021; however, it is important to note that the American economic, political, and social climate has changed significantly in 2020 due to the COVID-19 pandemic, social and racial injustice protests, the 2020 presidential election and other major events. While these factors are not a part of this research project, we acknowledge that these factors are likely to change the total amounts donated, to which charities some of the gifts are donated, and who donates. We analyze five proposed changes to the federal income tax to provide incentives to non-itemizers for charitable giving (Indiana University Lilly Family School of Philanthropy 2019b). The policies are described in the data and methods section. Key results show that four of the five policy proposals bring in more charitable dollars than are lost in Treasury revenue; four of the five policy proposals bring in more charitable dollars than were projected to have been lost as a result of TCJA; and all five proposals bring in more donor households that were expected to be lost as a result of TCJA.

Background
Charitable giving is an important part of American life (de Tocqueville 1835; Di Mento 2020). While there are many reasons that motivate people to donate, one influence is the potential tax benefit that they receive from their charitable contributions Wiepking 2011a, 2011b;Konrath and Handy 2018).
In the United States, individuals, including citizens, residents, and nonresidents, have been required to pay income taxes since 1913. Taxpayers must pay a certain portion of their taxable income to the federal government. The portion owed is determined by their income, legal deductions and credits, and their marginal tax rates. Their taxable income is determined by subtracting their tax deductions from their adjusted gross income (AGI). One such deduction is the charitable deduction, which was added in 1917. Currently, the charitable deduction is only available to itemizers (Urban-Brookings Tax Policy Center 2018). In addition, taxpayers may qualify for tax credits, which are subtracted from the taxes owed. Tax credits reduce a taxpayers' tax liability (amount owed; United States Internal Revenue Service 2019b), but there are two types of tax credits, refundable and non-refundable (United States Internal Revenue Service 2019b). If a credit is refundable and the tax credit is larger than the taxpayers' tax liability, the taxpayer The Impact of Five Different Tax Policy Changes on Household Giving will receive the remainder of the credit in a refund, whereas if it is non-refundable, the taxpayer cannot benefit from the credit above their tax liability (United States Internal Revenue Service 2019b).
Furthermore, tax deductions sometimes have additional limitations, including floors and caps. Floors set a minimum amount (dollar amount or percentage of AGI) below which the deduction cannot be claimed (Galle, Colinvaux, and Steuerle 2012;Lowry 2014) and caps set a maximum amount (dollar amount, percentage of AGI, or percentage of standard deduction) that can be deducted (Feldstein 2015;Galle, Colinvaux, and Steuerle 2012).
In addition, taxpayers can choose between the standard deduction or the total of their itemized deductions; this is typically referred to as itemization status. Itemizers' taxable income is reduced by the total amount of their deductions, while non-itemizers' taxable income is reduced by the standard deduction. In 2018, the standard deduction was $12,000 for single taxpayers and $24,000 for taxpayers who were married and filing jointly, which is nearly twice what it was in 2017. Table 1 shows the change in the standard deduction before and after TCJA was passed. The number of households who itemize declined significantly after the passage of TCJA, with 87 percent of taxpayers taking the standard deduction for tax year 2018 compared to 69 percent in 2017 (United States Internal Revenue Service 2019a, 2020). This means that approximately 28 million households were no longer be able to benefit from the charitable deduction in 2018 compared to 2017, including many middle-and upper-income taxpayers.
Recent trends, including the decreasing share of American households donating to nonprofits and the passage of the 2017 Tax Cuts and Jobs Act (TCJA) have led to concerns among nonprofit organizations that the overall increases in charitable giving might not be sustainable and that the increasing concentration of giving among high-income households could lead to greater inequality (Giridharadas 2018; Indiana University Lilly Family School of Philanthropy 2019; Reich 2006Reich , 2018. Some research (Brill and Choe 2018;Gleckman 2018;Ricco 2018; Tax Foundation 2017; Urban-Brookings Tax Policy Center 2017b), including a recent study researched and written by the school and commissioned by Independent Sector (Rooney et al. 2017), have projected that some of the provisions included in the new policy (e.g. the increase in the standard deduction, the cut in marginal rates, the continued decrease in the number of people subject to the estate tax, the cut in corporate rates, etc.) will significantly reduce charitable giving (Brill and Choe 2018;Gleckman 2018;Ricco 2018;Rooney et al. 2017;Tax Foundation 2017;Urban-Brookings Tax Policy Center 2017b). The long term effects of TCJA may not be known for years. However, there is some evidence of short-term effects. Giving by individuals decreased 4.6 percent adjusted for inflation 2018 from 2017despite significant growth in factors that are typically associated with growth in household giving (e.g., income and wealth) relative to the status quo ex ante. Giving by individuals increased in 2019 from 2018, but did not return to the all-time high of $315.82 billion given by individuals in 2017 (Giving USA Foundation 2020).
As a response to these concerns and as a way to offset the projected loss in charitable donations, various proposals have been made to extend the charitable deduction to non-itemizers. These proposals include the Charitable Giving Tax

Provisions of TCJA Affecting Charitable Giving
In addition to the standard deduction, other provisions of TCJA directly impact the funding of charitable organizations (e.g. donations) and the operational side of charitable organizations (Bailey 2017;Brant and Nofziger 2018). These policy changes vary with regards to whether they are expected to have a positive effect on charitable giving and nonprofits (e.g. the increased limit on cash contributions) or a negative effect on charitable giving and nonprofits (e.g. the increased exemption for estate and gift tax); changes to the calculation of taxes owed on Unrelated Business Taxable Income (UBIT; Independent Sector 2018). Furthermore, other changes in tax policy resulting from TCJA that do not directly affect nonprofit organizations are still likely to indirectly affect charitable giving, particularly for high income donors (e.g. $10,000 cap on the State and Local Tax (SALT) deduction). Space does not permit us to go into details here, so please see the full report.

The Impact of Five Different Tax Policy Changes on Household Giving
It is clear that federal tax policy is not only complicated, but the various provisions have both direct and indirect effects, as well as potentially positive and negative effects on charitable giving. Therefore, it is important to conduct research on how this and future policies could affect charitable giving and nonprofit organizations more generally.
The tax treatment of charitable giving influences who gives and how much they give (Feldstein and Taylor 1976). Between 2000 and 2016, the number of households that donate to nonprofits declined, and this trend is especially strong among low-and middle-income households (Clark, Han, and Osili 2019; Indiana University Lilly Family School of Philanthropy 2019a; Indiana University Lilly Family School of Philanthropy 2019). TCJA could also lead to a decline in donors. In addition, even though charitable giving dollars have been increasing in recent years (Giving USA Foundation 2020), this trend might not be sustainable due to the changes associated with TCJA as well as the past decrease in number of donors overall.
Because of this, nonprofits, policymakers, and academics have attempted to examine what other policy options could reverse negative giving trends. This report focuses on five policy options under consideration by the nonprofit sector. These options aim to offset the continued decline in donors, unequal treatment of taxpayers' charitable gifts in the tax code, and the potential decrease in charitable giving resulting from TCJA (Brill and Choe 2018;Ricco 2018;Rooney et al. 2017) by targeting non-itemizers and making the tax incentives for charitable giving more equitable. After a brief summary of the methods used to estimate the effects of the five policies, the paper will summarize the five policy options, present their estimated effects on charitable giving dollars, number of donor households, and Treasury revenue.

Data and Methods
Microsimulations to estimate the effects of various tax proposals on charitable giving were conducted using the Penn Wharton Budget Model (PWBM; Penn 2019b). This model is used to project the effects of government tax and spending policies and has been used in past research to predict the effects of policy changes on charitable giving (Reich 2018). The model uses a simulated population dataset that matches the overall population for many demographic and macroeconomic variables (Penn 2019a(Penn , 2020c. The PWBM was chosen for the following reasons: it is created by nonpartisan academic institution; the model is transparent in the methods it uses and in providing all results to the public regardless of the policy implications; and it combines two popular approaches used in economic models: reduced-form models typical of industry and public policy groups and structural models favored by academics (Penn 2020a(Penn , 2020b). In addition, in an analysis comparing seven overlapping-generations models (OLG models), all seven models, including the PWBM, provided similar results (Nelson et al. 2019 et al. 2017). There are two main benefits of using the income-based elasticities calculated using PSID data. First, the PSID includes giving by non-itemizers, while many other elasticities are calculated using tax return data provided by the IRS. Second, the PSID allows us to calculate different elasticities for each income group instead of relying on an overall estimated elasticity that does not vary by income group (Rooney et al. 2017). Additional information on the calculations to estimate these income-based tax-price elasticities of giving can be found in Supplemental Materials. The income-based elasticities were used in conjunction with the Penn Wharton Budget Model to estimate the effects of five policy changes on charitable giving dollars and number of donors, both overall and by income group (first quintile, second quintile, third quintile, fourth quintile, 81-90 percent, 91-99 percent, and top one percent). In all analyses, the charitable deduction for itemizers was not changed. In addition, limitations that apply to the itemizer deduction also apply to the non-itemizer deduction (e.g. 60 percent AGI limit on cash contributions).
Furthermore, because some previous research, including the Congressional Research Service and Congressional Budget Office (Congressional Budget 3 This paper analyzed the effects of a reduction in Social Security's Old-Age and Survivors Insurance benefits. While the analysis itself is significantly different from the analyses we conduct in our study, the purpose of the comparison in this paper is to test for agreement across models, making it relevant for our analysis. 4 $309.66 billion was donated in 2019 in current dollars. Adjusting for inflation, this is equivalent to $304.15 billion in 2018 dollars.

The Impact of Five Different Tax Policy Changes on Household Giving
Office 2011; Gravelle 2005), assumes lower responsiveness by taxpayers to changes in charitable giving, we also estimate the same analyses using a low responsive elasticity (−0.5) and a moderate responsive elasticity (−1.0; see Supplemental Materials).
In addition to outputs for charitable dollars donated and federal tax revenue, the PWBM was used to estimate the effects on the share of taxpayers who donate to nonprofits. To model how tax policy matters for people deciding whether or not to donate, the PWBM uses the average of the first-dollar tax price change and the next-dollar tax price change. 5 This measure provides a fuller picture of how tax policy matters for people deciding whether to donate, who likely think about both margins to some extent. Based on previous literature estimating extensive marginal elasticity, we use an elasticity of −0.8 for these analyses (Almunia, Lockwood, and Scharf 2017).
Finally, to understand the effects of each policy on charitable dollars and the decision to donate relative to the loss in Treasury Revenue, we calculated the net dollars, which is the difference between the amount of additional dollars donated to charity and the cost to the treasury, and relative donor incidence rates, which is the number of new donor households per $1 million in Treasury costs.
Using the TCJA as a baseline, we used the microsimulation models to estimate the effects of the following proposals: 1. A non-itemizer charitable deduction (Policy 1); 2. A non-itemizer charitable deduction with a cap for non-itemizers of $4000 for single filers and $8000 for married couples filing jointly (Policy 2); 3. A non-itemizer charitable deduction with a modified one percent floor that allows non-itemizers to deduct 50 percent of the value of their charitable gifts under one percent of AGI and a normal deduction for gifts over one percent of AGI (Policy 3); 4. A non-itemizer non-refundable 25 percent charitable giving tax credit (Policy 4); and 5. An enhanced non-itemizer charitable deduction, which provides a higher value deduction for low-and middle-income households (Policy 5): -Single filers earning under $20,000 can deduct 200 percent of the value of their charitable donations, single filers earning between $20,000 and $40,000 can deduct 150 percent of their charitable donations, and single filers earning over $40,000 can deduct 100 percent of their charitable donations; 6 and -Married couples filing jointly earning below $40,000 can deduct 200 percent of the value of their charitable donations, married couples filing jointly earning between $40,000 and $80,000 can deduct 150 percent of their charitable donations, and married couples filing jointly earning over $80,000 can deduct 100 percent of their charitable donations.
The results of these analyses, as well as explanations of each policy option are included in the results section. For each proposal we estimated the impact on charitable giving dollars, number of households that donate to nonprofits, and Treasury revenue.

Results
Under TCJA, the model predicts that 88.6 million households will donate approximately $342 billion in 2021 (see Table 2). In line with previous research In addition, all five policy options are estimated to more than offset the potential negative effects of TCJA on the number of households that donate. Four of the five policy options are estimated to more than offset the potential negative effects of TCJA on charitable giving and bring in more charitable dollars than are lost in Treasury revenue. The non-itemizer deduction with a $4000/$8000 cap is the only policy option that does not bring in more charitable dollars than were projected to be lost as a result of TCJA or than could be lost in Treasury revenue (see Table 2).
The 25 percent credit for non-itemizers (Policy 4) is estimated to have the largest effect on both charitable dollars donated and the number of households who donate. The credit could increase giving by $37 billion 7 (11 percent 8 increase) in 2021 with 10.6 million additional households donating (12 percent increase). However, it would also be the most expensive option for the Treasury, costing $33 billion in Treasury revenue (0.9 percent decrease).
The enhanced non-itemizer deduction (Policy 5) is estimated to have the next largest effect on charitable giving dollars and number of donor households, increasing charitable giving by $29 billion (nine percent increase) and increasing the number of households donating by 8.4 million (10 percent increase) households. It would reduce Treasury revenue by $24 billion (0.7 percent decrease), making it the second most expensive option for the Treasury.
The basic non-itemizer deduction (Policy 1) is estimated to have the third largest effect on charitable giving dollars and number of donor households, increasing charitable giving by $26 billion (eight percent increase) and increasing  7 All results presented are maximum possible effects under the conditions described in the "Data and Methods" section. 8 All percent changes presented are relative to current federal tax law at the time of analysis (TCJA).
the number of households donating by 7.3 million (eight percent increase) households. It would reduce Treasury revenue by $22 billion (0.6 percent decrease), making it the third most expensive option for the Treasury. The non-itemizer deduction with a modified one percent floor (Policy 3) is estimated to have the fourth largest effect on charitable giving dollars, but the smallest effect on number of donor households, increasing charitable giving by $25 billion (seven percent increase) and increasing the number of households donating by 4.6 million (five percent increase) households. It would reduce Treasury revenue by $18 billion (0.5 percent decrease), making it the least expensive option for the Treasury.
The non-itemizer deduction with a $4000/$8000 cap (Policy 2) is estimated to have the smallest effect on charitable giving dollars, but the second smallest effect on number of donors, increasing charitable giving by $17 billion (five percent increase) and increasing the number of households donating by 7.0 million (eight percent increase) households. It would reduce Treasury revenue by $20 billion (0.5 percent decrease), making it the second least expensive option for the Treasury. This is the only policy option that does not bring in more charitable dollars than are lost in Treasury revenue.
We created two "aggregate summary statistics" to try to assess "the best" option of these five. In Column D in Table 3 (below) we added the ranks of the five policies for their effects on charitable dollars and donors. Using these measures, the 25% tax credit is the "best" option, followed by the enhanced UCD, then the traditional UCD. If we include the forgone Treasury revenues, the rank order remains the same. While the enhanced UCD would be more progressive than a Table : Aggregate-summary statistic based on the ranks of effects of five policy options on charitable giving, number of donors, and Treasury revenues in .  Not only has this led to fears that the positive trend in charitable giving dollars is not sustainable (Osili and Zarins 2018;Rooney 2018Rooney , 2019, but it raises issues of equity among donors and nonprofits since giving is becoming increasingly concentrated among high-income households (Giridharadas 2018;Osili, Clark, and Bergdoll 2018;Reich 2018;Rooney 2018Rooney , 2019. Furthermore, the 2017 Tax Cuts and Jobs Act was one of the largest changes to federal income tax policy in recent history and it was expected to have a significant negative impact on charitable giving (Brill and Choe 2018;Gale et al. 2018;Ricco 2018;Rooney et al. 2017;Rosenberg and Stallworth 2017;Urban-Brookings Tax Policy Center 2017a). Nonprofit leaders and policymakers have been trying to find policy options that stimulate charitable giving (both total dollars donated and the number of households that donate) and are equitable. Over time, federal tax policy in the United States has become less progressive, lowering the tax burden on highincome households without providing the same reductions for low-income households (Piketty and Saez 2007). Recently, there have been various proposals by policymakers and industry research groups to change tax policy at both the federal and state levels to make the tax system more progressive (Grundman 2019). However, among nonprofit leaders, advocates, policymakers, and academics, proponents of a more fair and equitable tax code remain unclear about which giving incentives might best achieve their goals.
The five policy proposals analyzed in this report target tax reform for nonitemizers, which primarily includes low-and middle-income households. However, subsequent to TCJA, many high-income households are now non-itemizers.
While four of the five proposals could potentially offset the predicted drop in charitable dollars as a result of TCJA, a non-itemizer non-refundable 25 percent tax credit would have the largest positive impact overall, in addition to being one of the fairer proposals because it provides the same benefit to all taxpayers regardless of their marginal tax rate. For example, a $1000 donation generates a $250 tax credit regardless of one's income or top marginal tax rate. This tends not to be the case for other policies. Furthermore, all five proposals more than offset the predicted drop in in number of donor households as a result of TCJA, with the nonitemizer non-refundable 25 percent tax credit having the largest positive impact on the number of households that donate to nonprofits overall and at every income level except the top one percent.
While the non-itemizer non-refundable 25 percent tax credit has the biggest positive impact on gross charitable giving dollars and number of donor households, it is also the most expensive option for the Treasury. Therefore, it is important to consider which policy has the largest net positive impact on charitable dollars and the number of donor households compared to the cost to the Treasury. The non-itemizer deduction with a modified one percent floor has the largest positive net impact on charitable dollars, bringing in an estimated $7 billion more in charitable dollars than are lost in Treasury revenue (in other words, this policy brings in $1.39 for each $1 lost in Treasury revenue). The non-itemizer deduction with a $4000/$8000 cap has the largest positive impact on the number of donor households relative to Treasury costs; it would induce an additional 352 households to donate to nonprofits for each $1 million lost in Treasury revenue. When considering the aggregate-summary statistics based on the ranks of effects of the five policy options, the non-itemizer non-refundable 25 percent tax credit ranks highest (i.e., lowest sum of ranks) both when only considering the philanthropy variables (aggregate-summary statistic is 2) and when incorporating the philanthropy variables and Treasury revenue (aggregate-summary statistic is 7).
There are many issues to consider when examining the impact of tax policies. Not only should nonprofit leaders and advocates as well as policymakers consider the effect of each policy on charitable giving dollars, the number of households that donate, and Treasury revenue, but they should consider issues of donor equity and efficiency. This report is meant to provide empirical data to inform conversations surrounding incentives for charitable giving for non-itemizers.
While this study has many strengths, there are some limitations that should be noted. First, policy options may change. This study attempts to analyze the most relevant policy options available. However, these policies could change as the debates among nonprofit leaders, advocates, and policymakers continue. Not only can these policies change prior to their introduction as bills in the United States Congress, but those that are or have been introduced as likely to be changed during negotiations throughout the relevant committee, House, and Senate debates. Second, researchers don't fully understand the magnitude of TCJA on charitable giving. TCJA was a significant change to federal tax policy. While a lot of research is being done to both predict and measure the effects, the actual effects are still unclear and won't be fully understood for many years. In addition, the literature shows that research on the effects of tax policy varies based on the dataset used. Therefore, while we used the best available data, it is possible that other datasets could produce different results. Furthermore, there are some considerations that are important when considering changes to tax policy that were not measured or estimated as part of this study. For example, we did not estimate the administrative or the compliance costs of implementing the individual policies. The effects of the various policies could vary by region or state, 9 which is also an important consideration. Finally, the forecasts were made in early 2019. Given the major economic, social, political, and health changes in 2020 (Willingham 2020), the results may not be as relevant today as they were in 2019. However, the results still provide useful information for researchers, nonprofit leaders, and policymakers.
resulting from five policy proposals," which can be found at this link: http://hdl. handle.net/1805/19515 Research funding: This project was sponsored by Independent Sector (IS) with the explicit understanding that the Indiana University Lilly Family School of Philanthropy at IUPUI (IULFSOP) would retain the rights to present and publish the results-regardless of the outcomes and how those outcome might be perceived by various political or philanthropic communities. This research was possible thanks to the generosity of Charles Stewart Mott Foundation and Fidelity Charitable Trustees' Initiative.