Study Concludes TCJA Reduced Charitable Giving by $20 Billion Annually

A new study quantifies the effect of the Tax Cuts and Jobs Act on charitable giving:

ABSTRACT

The Tax Cuts and Jobs Act eliminated federal charitable giving incentives for roughly 20 percent of US income-tax payers. We study the impact of this on giving. Basic theory and our empirical results suggest heterogeneous effects for taxpayers with different amounts of itemizable expenses. Overall, the reform decreased charitable giving by about $20 billion annually. Using a new method to adjust estimates for retimed giving, we find evidence of moderate intertemporal shifts from pre-announcement of the law. The permanent price elasticity of giving estimates range from .6 for the average donor to over 2 for those predicted to be most responsive to the reform.

Here is a short excerpt from the Tax Incentives for Charitable Giving:  New Findings from The TCJA:

In this paper we present estimates of donor responsiveness to a charitable deduction based incentive that address these problems. The estimates are based on changes effected by the Tax Cut and Jobs Act of 2017. TCJA was a three trillion dollar policy, and created the largest change in US giving incentives in a generation. Starting in 2018, TCJA induced an extremely large group of taxpayers—roughly 20% of all households—to use the standard deduction instead of itemizing. These taxpayers thus lost their US-tax-code incentive to give. Our first contribution is to exploit the fact that this reform created large changes in incentives even for taxpayers who experienced little change in income and marginal tax rates. This allows a study of a shock in the US tax code that addresses both identification problems inherent in studying deductibility.

Second, we demonstrate theoretically that the effects of TCJA on giving are heterogeneous across taxpayers, determined by their pre-TCJA level of itemized deductions. For some taxpayers the predominate TCJA effect is a large increase in post-tax income, which could increase giving. But taxpayers whose pre-TCJA deductions were close to the new TCJA standard deduction experienced TCJA as a compensated price increase. For these taxpayers we can estimate a compensated price elasticity. Knowing the compensated price elasticity obviously is important for optimal tax policy and welfare analysis. This estimate is the first of a compensated elasticity for total charitable giving.

Third, we introduce a new approach, based on the models in Auten, Seig, and Clotfelter (2002) and Bakija and Heim (2011), to address the intertemporal shifting confound to estimates of permanent price elasticities. The central intuition is that intertemporal shifting generates autocorrelation between anticipatory spikes and subsequent drops in giving. The new approach is easy to implement, and applicable in other contexts in which a policy change is pre-announced.

We estimate the effect of TCJA on giving using data from the Panel Study of Income Dynamics. The data describe 4,232 Americans over 2000-2018. Our approach to identification is to compare the change in giving among taxpayers predicted to stop itemizing as a result of TCJA to the change in giving among taxpayers not predicted to switch itemization status. The policy effect can be identified while using extensive controls to ensure that taxpayer-specific variation in income does not confound the estimates.

The baseline intention-to-treat estimate is that TCJA caused giving to drop by $366 per taxpayer predicted to switch. The treatment-effect-on-the-treated estimate suggests an $880 drop in giving per taxpayer who actually switched. The policy decreased charitable giving by about $20 billion annually relative to trend; this result suggests the entire observed drop in aggregate giving in 2018 was due to TCJA.

Applying the new approach to intertemporal shifting using different datasets and assumptions all produce a similar conclusion: about 20 percent of the post-TCJA drop in giving reflects gifts shifted intertemporally. After accounting for intertemporal shifting, the estimates imply a permanent price elasticity of .6. Furthermore, there are heterogeneous responses across taxpayers in line with theoretical predictions. For taxpayers facing relatively large positive income effects from the law, we find insignificant or in some cases positive effects of TCJA on giving. The overall decrease in giving is driven by those predicted to experience TCJA as a compensated price increase; for these taxpayers the compensated price elasticity is over 2.

As a fourth contribution, we break down the results by type of charitable donation. The results indicate policy-relevant heterogeneous responses across different types of giving. TCJA caused little change in giving to religious congregations. Instead, the decrease was almost entirely in giving to organizations with other purposes, and among them a large portion of the drop was to organizations that help people in need. Finally, the results have an important implication for interpreting the divergence of donor responsiveness estimates in the literature. The present results uncover significant heterogeneity in responsiveness across taxpayers and across types of nonprofit organizations using a single dataset, one policy shock, and a single methodological approach. Divergent estimates on the responsiveness of donors can represent true heterogeneity. We return to this implication in the conclusion.

darryll k. jones