Polarizing the Donorate: Charitable Giving and Negative Political Advertisements

ABSTRACT

Does inflammatory political rhetoric undermine the voluntary provision of public goods? This paper addresses this question by combining Form 990 data on charitable giving with Wesleyan Media Project data on political advertising and a text-based measure of charity ideology. Attack advertising creates political polarization, which discourages giving to ideological charities, and encourages giving to non-ideological charities. These results suggest that divisive political rhetoric may actually strengthen nonpartisan civil society.

KEYWORDS:

• Altruism
• charitable giving
• political advertising
• political polarization

Acknowledgments

I am grateful for comments from Benjamin Marx, Marco Castillo, and seminar participants at the Michigan Tax Invitational, the Science of Philanthropy Initiative, the Cornell Law and Economics Workshop, the National Tax Association Annual Meeting, and the Allied Social Science Associations Annual Meeting. I would also like to thank Elliot Ash and seminar participants at the Monash-Warwick-Zurich Text-as-Data Workshop for invaluable feedback concerning the construction of the ideology measure. I gratefully acknowledge financial support from the Office of Tax Policy Research at the University of Michigan in obtaining data for this project.

Disclosure statement

This research was conducted while the author was an employee at the U.S. Department of the Treasury. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors and do not necessarily reflect the views or official positions of the U.S. Department of the Treasury. Any taxpayer data used in this research was kept in a secured Treasury or IRS data repository, and all results have been reviewed to ensure that no confidential information is disclosed.

Data availability statement

The data were obtained from the Wesleyan Media Project, a collaboration between Wesleyan University, Bowdoin College, and Washington State University, and includes media tracking data from Kantar/Campaign Media Analysis Group in Washington, D.C. The Wesleyan Media Project was sponsored in 2012–2018 by grants from The John S. and James L. Knight Foundation, and in 2012–2014 by grants from The John D. and Catherine T. MacArthur Foundation. The opinions expressed in this article are those of the author and do not necessarily reflect the views of the Wesleyan Media Project, Knight Foundation, MacArthur Foundation or any of its affiliates.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/10495142.2024.2360192

Notes

1. All instrumental variables specifications will be estimated using the generalized method of moments.

2. Under this alternative mechanism, donors think about charitable giving and voting as substitutes (Gee & Meer, 2019). If this hypothesis were to hold, then negative political advertising could affect private charitable giving through its impact on voter turnout (Ansolabehere et al., 1994; Lau & Rovner, 2009; Fraga & Hersh, 2010). Online Appendix A explores this hypothesis, using instrumental variables related to inclement weather on Election Day (Gomez et al., 2007; Lamare, 2013; Zelin & Smith, 2022). The results rule out evidence of a strong negative relationship between voting and charitable giving, though they allow for the possibility that weaker forms of this relationship exist.

In the long run, attack ads may reduce lawmakers’ willingness to compromise, thereby lowering public good provision by the government and prompting private donors to step in and fill the gap. This mechanism is related to traditional notions of crowd-out (Bergstrom et al., 1986; Warr, 1983). However, it is not obvious that changes to the volume of lawmaking will inspire more charitable giving. This requires charitable donors to have a high degree of awareness about the subject and pace of lawmaking, as well as the amount of government spending this lawmaking will create. Horne et al. (2005) find that charitable donors’ awareness of government support for charities is very low; more recently, Roth et al. (2022) find that over 90% of their nationally representative sample underestimates the true debt-to-GDP ratio. The plausibility of this alternative mechanism therefore depends heavily on an assumption of salience that is unlikely to be met.

3. Private charitable contributions are defined as the difference between the charity’s total contributions, found in Part VIII, Line 1 h, and the charity’s government grants, found in Part VIII, Line 1e.

4. See Form 990, Section VI, Line 17 for tax years 2012–2017.

5. These states include Arizona, Delaware, Idaho, Indiana, Montana, Nebraska, South Dakota, Vermont, and Wyoming.

6. In using this set of responses, this paper measures ideological polarization, as opposed to affective polarization, which is the extent to which members of different political groups feel negatively toward one another (Iyengar & Westwood, 2015; Iyengar et al., 2012, 2019). However, the questions necessary to construct a measure of affective polarization at the state-year level come from surveys fielded only once every four years, as opposed to annually. This poses a challenge to measuring affective polarization in the present study, which spans only six years. However, other studies have shown that in recent years, ideological polarization and affective polarization bear a strong, positive relationship to one another (Rogowski & Sutherland, 2016; Webster & Abramowitz, 2017). In the absence of a true measure of affective polarization, a measure of ideological polarization should prove a reasonable proxy.

7. When this assumption is relaxed, the results are broadly unchanged.

8. This is found in Box J.

9. This figure is expressed as a share of all charities for which a continuous measure of ideology can be produced; that is, it omits the 30,647 charities which either reported no URL in Box J, or for which the scraper was unsuccessful at retrieving a body of text.

10. These data are reported in Table 19 prior to 2015, Table 12 in 2015, and Table I-16 from 2016 onwards.

11. Karol (2024) notes that political activity can have profound effects on charities’ fundraising expenses. These fundraising expenses are very influential in determining private charitable giving. Online Appendix C re-estimates all specifications in this section, controlling for fundraising expenses. As fundraising expenses and private contributions are simultaneously determined, this control variable requires an instrument. Following Andreoni and Payne (2011) and Heutel (2014), the inverse hyperbolic sine of office expenses are used as an instrument for the inverse hyperbolic sine of fundraising expenses. The inverse hyperbolic sine (IHS) transformation is used to retain observations in which the charity reported $0 in fundraising expenses or office expenses. The results are broadly consistent with those presented in the main text.

12. Online Appendix D explores how the relationship between political attack ads and charitable giving depends on the polarization of the environment. This appendix splits the sample into two parts, according to whether or not the charity’s market scored above the median value for the RQ index at the beginning of the study period. In markets with an above-median level of initial polarization, attack ads benefit moderate charities and do not harm polarized charities. In markets with a below-median level of initial polarization, attack ads harm polarized charities and do not affect moderate charities. These results are consistent with the disgust hypothesis. In a low-polarization environment, increases in toxic political communication cause donors to turn away from ideologically extreme charities. When donors are already polarized, increases in toxic political communication inspire them to turn toward ideologically moderate charities.

Additional information

Funding

The author has no relevant financial or non-financial competing interests to report.