______
Religion and Gun Purchasing amid a Pandemic, Civil Unrest, and an Election
ABIGAIL VEGTER AND DONALD HAIDER-MARKEL
As the severity of the COVID-19 pandemic became clear in the spring of 2020, American debate over the freedoms granted by the First and Second Amendments became a nearly daily part of the national conversation. Concerns over empty store shelves and mixed messages from public officials highlighted concerns about personal safety for some. Likewise, restrictions on gatherings during local and state shutdowns, including restrictions for gun shops and religious services, led some to claim that their rights were being unconstitutionally violated. Even with largely ample access to firearm purchasing, many worried that the sales of guns would become limited. The NRA sued the state of New York and others who declared gun stores “nonessential” (Orden 2020). Likewise, some politicians argued that pandemic restrictions were subversive threats to freedoms. President Donald Trump tweeted his support for the “Reopen America” and “Liberate Michigan” campaign in April, linking the shutdowns to gun rights (Shear and Mervosh 2020). Texas attorney general Ken Paxton stated that emergency stay-at-home orders could not be used to close firearm stores in the state (Goldenstein 2020). As churches and gun stores shut down as a result of COVID-19, individuals, activists, politicians, and pastors alike claimed serious threats to First and Second Amendment rights. These claims, however, did not come to fruition as the pandemic has ushered in the highest rates of gun purchasing since such data began being tracked in 1998. A record breaking 3.9 million firearm background checks were completed in the month of June 2020 alone, and a total of nearly 39.7 million firearm background checks were completed in 2020, beating the previous yearly record by more than 10 million (Federal Bureau of Investigation 2020).1 Moreover, the Firearm Industry Trade Association estimated that 40 percent of sales were conducted with purchasers who had never previously owned a firearm (NSSF 2020).
To some, the pandemic restrictions also threatened their First Amendment freedom of religious practice. Political and religious leaders around the country opposed various restrictions on group gatherings and religious services. Some religious leaders ignored public officials’ orders and held services despite potential legal ramifications (Jacobo 2020) even as gatherings for religious services became noted super-spreader events for the virus (Conger, Healy, and Tompkins 2020).
In addition to the threat of the pandemic, Americans also faced a reckoning with police violence against African Americans. The murder of George Floyd while in police custody in May spurred widespread protests and occasional violence across the country for much of the late spring and summer. For some Americans, the unrest provided another layer of fear and uncertainty in the midst of an election that was argued to be a “Battle for the soul of our nation” (Cummings 2020). Then president Trump explicitly linked religion and civil unrest by having peaceful protesters tear-gassed and forcibly removed from Lafayette Park in order to walk across to St. John’s Church for a photo op of himself holding a Bible (“Peaceful Protesters Tear-Gassed to Clear Way for Trump Church Photo-Op” 2020).
As the pandemic raged and civil unrest and the unusual presidential election campaign continued, religion and guns became more intertwined. In addition, recent research suggests that some religious Americans justify their gun ownership using their faith. Vegter and Kelley (2020) find that gun ownership is part of a particular worldview that stems from a posture of fear and a commitment to a higher authority above the law. This particular “ethic,” as they describe it, relies on a particular understanding of Christian duty, including the duty to defend. The subjects from whom gun owners feel an obligation to defend themselves and their communities can include Democratic leaders, violent protesters, and, for some, Satan himself. The pandemic presents another set of fears requiring defense, especially concerning individuals’ health and safety.
How do these distinct expressions of fear, duty, and divine protection relate to the surge in gun purchasing during the turbulent summer of 2020? To examine this question, we deployed an online survey of a representative sample of American adults in October 2020. Our analysis of the resulting data suggests that evangelicalism significantly predicts the purchasing of a firearm between the months of March and August 2020. Moreover, this relationship varies by religiosity; those who claim that religion is very important in their lives are more likely to have purchased a gun during the spring and summer of 2020. When asked why they purchased a gun, evangelicals were significantly more likely to claim it was due to concerns about civil unrest as a result of police actions toward African Americans and the coming election rather than the threat of COVID-19. In other words, purchasing a gun resulted from a fear of civil unrest and that a Democratic president would “take their guns” rather than the pandemic. Notably, religious importance intensifies this relationship, with the most religious individuals being most likely to purchase a firearm between March and August 2020.
Religion and COVID-19
Religion has been at the center of the COVID-19 pandemic. Religious gatherings have operated as “superspreading events” (Starr 2020), and religious individuals express their opposition toward COVID-19 restrictions in explicitly religious language (Fowler 2020). Polls have shown that evangelicals and those with higher levels of religiosity are more likely to distrust scientific sources about guidance during the pandemic (Burge 2020) and less likely to take recommended health precautions (Perry, Whitehead, and Grubbs 2020a). Meanwhile, Americans who are less religious or not religious at all were more likely to social distance, wear masks, and adhere to the recommendations provided by health experts (Hill, Gonzalez, and Burdette 2020).
Perry, Whitehead, and Grubbs (2020a) provide a robust explanation for the religious response to the global pandemic. The scholars turn to Christian nationalism to explain why some Americans engaged in “incautious behavior” and took fewer precautions than others. The scholars note that Christian nationalism (“an ideology that idealizes and advocates a fusion of American civic life with a particular type of Christian identity and culture”) has been “shown to lower Americans’ trust in science and scientific expertise; promote a view of (conservative Christian) Americans as God’s chosen, divinely protected people; bind them to siding with Trump; and likely reject information put forth by mainstream news media” (2020a, 406). These side effects of Christian nationalism help explain why so many Americans refuse to adhere to COVID-19 safety guidelines.
Additionally, Perry et al. (2020a) find that religious commitment influenced Americans in the opposite direction of Christian nationalism. More devout Americans were more likely to engage in precautionary behaviors. However, it should be noted that this result was found after controlling for Christian nationalism, showing an effect only among the small number of religious Americans who are not Christian nationalists. The evangelical reaction to COVID-19 and the scholarly work illuminating potential causal mechanisms for such findings align well with what the literature suggests about religion’s relationship to guns in the United States.
Religion and Guns
Religion has consistently been shown to be a significant predictor of gun ownership in the United States. Young (1989), Little and Vogel (1992), and Cox, Navarro-Rivera, and Jones (2013), among others, have noted that over time Protestants are more likely than members of other religious traditions to own guns and less likely to support gun control; however, few studies move beyond the broad religious category of “Protestant” into other measures of religion, including evangelicalism or religious importance, despite both variables being powerful explanations of political thinking and behavior (Hertzke et al. 2018).
Yamane (2016) concludes that “Protestant” as a measure of broad religious affiliation is irrelevant to gun ownership, controlling for other factors. However, he does find that evangelical Protestants exhibit relatively high levels of personal handgun ownership. Vegter and den Dulk (2020) find that evangelical Protestantism is significantly associated with a gun owner identity as compared to mainline Protestants, which holds important implications for gun control policy attitudes. Moreover, some scholars have suggested that the link between white evangelicalism and gun ownership, as well as opposition to gun control, results from an individualist impulse that emphasizes personal responsibility (including, presumably, self-protection) and the role of civil society rather than the state in addressing the root causes of violence (Hempel, Matthews, and Bartkowski 2012; Merino 2018). The interaction of religion and other cultural factors, including the nexus between religion, residence in rural areas, and hunting, are also common in the literature (Young 1989). Taken together, these studies are consistent with broader arguments that attribute patterns of gun ownership and attitudes about guns to tensions between individualist and collectivist cultural traditions in the United States (Celinska 2007).
What are the causal mechanisms that would explain the effects of religious affiliation or religiosity on gun ownership? Some studies have provided a primarily culture-based explanation of these effects. Young and Thompson (1995) examine support for punitiveness and find that fundamentalism is associated with civic punitiveness among whites, which is in turn associated with gun ownership among whites. Among Blacks, fundamentalism is associated with religious punitiveness and, subsequently, gun ownership. Yamane (2016), however, reports that the influence of religious characteristics on personal handgun ownership is not strongly mediated by punitiveness when civic punitiveness is assessed among all racial groups.
Another approach draws from the literature on social capital and especially social trust. Matthews, Johnson, and Jenks (2011) argue that religiosity is associated with higher levels of generalized social trust and, therefore, less fear of certain types of crime. Hempel, Matthews, and Barthowski (2012) come to a different conclusion, at least about a subset of the faithful. They find that theological conservatism, as a distinctive belief structure, is itself associated with lower levels of generalized trust. One may consider, then, that active religious participation leads to a particular form of trust that reduces the likelihood of gun ownership, unless that participation is within the confines of a belief system that encourages social distrust. While Matthews et al. (2011) or Hempel et al. (2012) suggest these possibilities, neither study addresses gun ownership directly.
Whitehead, Schnabel, and Perry (2018) argue that Christian nationalism is associated with decreased support for gun regulation. They explain that Americans who subscribe to this particular worldview may believe “guns are a God-given right tied to a cultural style tied to deeply held senses of morality, identity, and perceived threat” (9). The particular religious reasoning used to justify gun ownership becomes especially significant when one considers the gun purchasing surge experienced in the spring and summer of 2020.
Gun Purchasing in 2020
Firearm background checks have been a consistent way to measure gun sales in the United States since 1999. With background checks reaching record highs throughout the COVID-19 pandemic, scholars have been interested in assessing this particular “gun-buying” event in relation to others in recent history. Such events include elections, terrorist attacks, and mass shootings (Liu and Wiebe 2019). In the case of the turbulent year of 2020, there was an unprecedented pandemic, a national election, and a wave of nationwide protests for racial justice in the light of police treatment of African Americans. The literature has investigated which types of events trigger gun sales, though, notedly, has not investigated what happens when these events happen simultaneously.
Lang and Lang (2020) smartly frame the COVID-19 pandemic as a threat to safety that may trigger firearm sales similar to terrorist attacks and mass shootings. The scholars find the pandemic, as a particular gun-buying event quantified by background checks, had a greater impact on firearm purchases than any other event since 1999. In their multivariate analysis, Lang and Lang (2020) even find that the effects of the COVID lockdown and the George Floyd protests on the increase of firearm purchases was even greater than that of the 9/11 attacks.
Lang and Lang (2020) draw a clear line between COVID and civil unrest as perceived serious threats to personal safety that created the observed historic gun buying of 2020. Indeed, safety concerns have been shown to influence firearm purchasing in the past (Levine and McKnight 2017). Not only do the majority of gun owners cite personal protection as their main justification for owning a gun (Azrael et al. 2017), but scholars have also found that past victimization of crime and fears of future victimization are significant predictors of gun ownership (Hauser and Kleck 2013; Kleck et al. 2011). Lyons et al. (2020) find that the main motivation for gun purchasing between January and May 2020 was protection against people, suggesting threats to safety in myriad forms matter to firearm purchasing.
Political uncertainty has also been shown to influence increases in firearm purchases. Depetris-Cahuvin (2015) finds that gun purchasing increases during election years, as future gun policy is unknown. The 2008 and 2012 election and reelection of Barack Obama were shown to increase firearm purchasing (Laqueur et al. 2019), though this was not true after Trump’s 2016 election (Smith 2020). This finding aligns with those of LaPlant, Lee, and LaPlant (2021), who note that gun sale spikes are positively associated with Democratic presidencies, which was predicted throughout the spring and summer of 2020.
The death of George Floyd at the hands of police in May 2020 ignited national protest over police treatment of African Americans in the United States. Lang and Lang (2020) find that the George Floyd protests were associated with an increase in firearm purchasing, distinct from the COVID-19 outbreak. Moreover, the scholars find that there is not a partisan dimension to this increase, with Democrat-leaning and Republican-leaning states experiencing similar patterns, a finding they explain by noting both events do not involve gun policy uncertainty. As mentioned, gun sales often spike when gun policy seems uncertain and a Democratic presidency is perceived as a potential threat to gun rights (LaPlant, Lee, and LaPlant 2021), suggesting that there was something distinct about the events of 2020. Violent crime increased that summer (Federal Bureau of Investigation 2020) and a common narrative attributes the rise in violence to protests against police treatment of African Americans (King 2020). While evidence is inconclusive on how the protests influenced a rise in crime, Lyons et al. (2020) find that a major motivation for gun purchasing between January and May 2020 was fear of crime.
Indeed, considering the potential perceived threat of the COVID-19 pandemic, civilian unrest over police violence, and a contentious election year, we would expect the demand for firearms was high. However, which of these forces, if any, might have had more of an influence on demand? We can gain some insight to this question by examining web searches in 2020.
If we consider that a perceived personal and familial threat from COVID lockdowns and civil unrest might motivate gun purchases, we should expect that interest in guns for personal and family protection would increase. Coinciding with some Protestant Christian traditions, the need for individuals to defend against threats in the twenty-first century might best be resolved by owning a gun. Meanwhile, if the 2020 elections were perceived as potentially creating a change in gun policy, we should expect an interest in guns to be connected to interest in the gun policies of candidates. For conservative Christians, who have aligned themselves with the pro-gun-rights Republican Party, the potential threat of Democratic victory in November might increase the perceived need to purchase a first-time purchase of a gun or add guns to an existing collection (Djupe 2019).
We attempt to break down 2020 interest in guns by examining web searches for guns with Google Trends data. Here we used a search term of “gun for home” as a means of examining gun purchase interest for personal protection. We contrast those searches with searches under “gun policy” as a means to capture interest in the elections and gun positions of candidates. Our effort is similar to that used by Lang and Lang (2020). In Figure 3.1 we display Google Trends results for web searches in the United States using the phrases “gun for home” versus “gun policy” between December 2019 and December 2020. The search volume for each phrase is normalized over the series on a range of 0 to 100 based on the included phrase’s proportion to all searches.
The figure reveals a number of peaks in searches for each phrase, with “gun for home” having its highest peaks in March as most of the country went into COVID-induced lockdown and in late May and early June, at the peak of civil unrest over the police killing of George Floyd. Meanwhile, the peaks for “gun policy” searches occur in early 2020 while the Democratic presidential nomination was still contested, again in August as each party held its nominating convention, and in October and November—the peak of attention to the elections. Clearly interest in guns shifted throughout the year. The searches indicate that gun purchasing demand earlier in the year might have been influenced by those looking to buy firearms for personal and family protection, perhaps by first-time gun owners, while gun-related searches later in the year might have been motivated by gun owners and potential gun owners examining the gun policy positions of candidates. We can take a more refined look at these explanations, as well as assess the impact of religion on gun purchasing, using individual-level survey data.
Figure 3.1 Google Trends U.S. Web Searches for Key Terms, December 2019 to December 2020 (Source: Data compiled from Google.)
Data and Methods
For our analysis of individuals, our data come from an online survey we commissioned that was administered in October 2020 by Dynata. The survey sample was not a random probability sample of American adults but is representative of the adult population on all major demographic characteristics. Adults were recruited by Dynata, who invited participants via email to complete the survey (see the online appendix). The sample includes 1,784 adults. Of these participants, 513 identified as gun owners, and 150 of these gun owners indicated they purchased a gun between March 2020 and August 2020. Based on this, we estimate that roughly 8 percent of Americans (perhaps over 16 million adults) bought a gun during this time. About half of those who bought a gun during the time frame were first-time gun buyers (42 percent), which suggests that perhaps over 8 million American adults bought a gun for the first time between March and August 2020.
Variable Measurement
Our dependent variables consist of a measure for gun purchases made between March and August 2020 (“Did you buy a gun anytime between March 2020 and August 2020?”) and three measures assessing the rationale for that purchase: concern about COVID-19, concern about civil unrest because of police actions toward African Americans, and concern about the outcome of the 2020 election. The gun purchase question is simply dichotomous (0 = no; 1 = yes). The questions on motivation for the purchase allowed respondents to indicate “extremely important” (coded 6) to “not at all important” (coded 1) for COVID-19 concern, unrest concern, and election concern. We model gun purchasing with logistic regression and responses to each concern question with ordered logit.
Our chief independent variables include several measures of religious identification and religiosity. The survey includes the standard question “Would you describe yourself as a ‘born-again’ or evangelical Christian, or not?” Burge and Lewis (2018) find almost no statistical difference between the full, detailed religious affiliation approach and the simpler scheme that we utilize here. The literature is clear that there is something distinct about white evangelicals (Emerson and Smith 2020; Tranby and Harmann 2020), and therefore, we create a dummy variable for white evangelical identification, with all other respondents as the reference category. We investigate both measures of evangelicalism and white evangelicalism, recognizing the political relevance of the latter category. We additionally consider religious importance in a respondent’s life using the following question: “How important is religion in your life?” Responses were coded as 1 for “not at all important” to 4 for “very important.”
We include several variables to control for the demographic considerations shown to be important in research about guns and gun policy (Geier, Kern, and Geier 2017; Goss 2017; Kleck and Kovandzic 2009; Morin 2014; Pederson et al. 2015; Spitzer 2012). We control for race and gender through dichotomous variables in each model (white = 1; female = 1), as well as age (in years), education (on a 7-point scale), and partisan identification of respondents using a 7-point scale, in which strong Democrats are indicated by the highest value.
Results and Discussion
At the bivariate level, evangelicalism is positively associated with buying a gun between the months of March and August 2020 (r = 0.24, p < 0.001). While this may not initially seem like a particularly strong correlation, it is larger than the correlations for every other variable we consider, including religiosity (r = 0.23, p < 0.001), race (r = –0.046, p > 0.05), gender (r = –0.085, p > 0.05), and partisanship (r = –0.047, p > 0.05). Shown another way, 43 percent of evangelical Protestants in our sample bought a gun between the months of March and August 2020.
In the multivariate analysis (see full results in the appendix, Table A3.1), religiosity (β = 0.384, p < 0.01) and evangelicalism (β = 0.461, p < 0.05) are the leading predictors for buying a gun between the months of March and August 2020. When we interact evangelicalism and race, comparing white evangelicals to all others in the analysis, white evangelicalism loses predictive power, but religious importance remains significant (β = 0.384, p < 0.01). Women were less likely to purchase a gun during those months in both models (p < 0.05). When assessing the predicted probabilities for white evangelicals and others purchasing a firearm between the months of March and August 2020, all individuals who score higher on the religiosity scale experienced an increased likelihood of gun purchase, regardless of religious identity, though white evangelicals are especially impacted by religiosity (see full probabilities in the appendix, Figure A3.1).
Figure 3.2 presents the multivariate models assessing the reasons individuals purchased a firearm between the months of March and August 2020, including models of concerns over COVID-19, concerns over the upcoming 2020 election, and concerns over civil unrest due to police actions toward African Americans (see the full model in the appendix, Table A3.2).
Looking at Americans’ citation of concerns over COVID-19 as being an important factor in their decision to buy a gun, the more religious respondents were significantly more likely (p < 0.001) to indicate high importance in both the model with evangelicalism broadly and white evangelicalism more specifically. Surprisingly, neither evangelicals nor white evangelicals were significantly more or less likely to assign high importance to COVID-19 (p > 0.05). Our other variables do not hold significance in this exploration of COVID-19 as an important rationale for gun purchasing.
Evangelicals were significantly more likely to assign importance to concerns over the upcoming election in purchasing a firearm between the spring or summer of 2020 (p < 0.10). White evangelicals, however, do not have a significant relationship with this particular rationale. In addition, more religious respondents were more likely to indicate the elections as of high importance in their firearm purchase (p < 0.05) in both models. Other variables do not reveal any predictive power.
Figure 3.2 Estimated Effects on the Three Rationales for Gun Purchases (Source: KU 2020 survey.)
Model 3 investigates the role of concerns about civil rest over police actions toward African Americans in the purchasing of a firearm. White evangelicals were significantly more likely to cite concerns over civil unrest (p < 0.05), though importantly not evangelicals broadly, suggesting a racialized reaction to civil unrest as a motivator for gun purchasing. Religious respondents were also likely to utilize this rationale in both models (p < 0.05); meanwhile, Democrats were less likely to assign high importance to concerns over civil unrest in their decision to purchase a firearm in both models (p < 0.05).
Conclusion
Gun sales in the United States have sharply risen throughout the COVID-19 pandemic (Lyons et al. 2020). Americans also witnessed widespread civil unrest over police violence and a contentious election in 2020, perhaps motivating some to consider purchasing firearms. Given the exciting literature on religion and guns, we considered the role of religion in gun purchasing during 2020 and what events were more important for those American who did purchase a firearm. Our analysis of data from a representative sample of American adults allows us to draw several tentative conclusions.
First, based on commonly used measurements of gun purchases, gun buying in 2020 was historic (Lang and Lang 2020). Based on the results of our survey sample of 1,784 American adults, we estimate that about 8 percent (perhaps 16 million adults) purchased a gun between March and August 2020. About half of those that bought a gun during the time frame were first-time gun buyers (42 percent), which suggests that over 8 million American adults bought a gun for the first time between March and August 2020.
Second, our analysis suggests that evangelicals were more likely to purchase firearms in 2020. In addition, those with higher religious commitment were more likely to purchase a gun. Given the linkages found between religion and gun ownership, this finding is consistent with existing literature (Yamane 2016) and suggests that the pandemic did not change this previously observed relationship.
Third, we asked those who did purchase firearms whether concern over events in 2020 influenced their decision to buy a gun. Evangelicals were more likely to say their purchase resulted from concerns surrounding the upcoming election, with white evangelicals suggesting the civil unrest relating to police treatment of African Americans motivated their purchase. Neither evangelicals broadly nor white evangelicals specifically cited concerns about the pandemic. This suggests that evangelicals perceived events in 2020 differently than did others who purchased guns.
Evangelicalism’s individualist cultural tendencies, as opposed to a collectivist perspective (Emerson and Smith 2001), help explain evangelical gun purchasing in 2020. The emphasis on the individual involves a particular self-protectionist approach to unrest. When considering how evangelicals, perhaps especially those who adopt a Christian nationalist ethic (Perry, Whitehead, and Grubbs 2020a), respond to the COVID-19 pandemic, it is clear that the belief that America is uniquely protected by God precludes the need for self-protectionist action (i.e., purchasing a gun). However, the other forms of unrest experienced during the spring and summer of 2020, including protests over police violence toward African Americans and uncertainty about the outcome of the 2020 presidential election, apparently do require action to protect oneself and one’s way of life. Exercising a God-given right to purchase a firearm is a seemingly meaningful way to protect the distinct culture that white evangelicals subscribe to—one in which gun ownership is a natural continuation of an emphasis on the individual (Celinska 2007).
Fourth, among religious gun purchasers, concerns about the global pandemic and the 2020 election appear to have played a greater role in motivating their purchase than did civil unrest. In light of this result, it does appear that religious importance has a unique relationship with the justification for purchasing a firearm, one that is distinct from evangelicalism. Unlike evangelicalism broadly or white evangelicalism specifically, religiosity is positively and significantly associated with the expression that concern over COVID-19 was important in the decision to buy a gun. Matthews, Johnson, and Jenks (2011) argue that religiosity is associated with higher levels of generalized social trust and, therefore, less fear of certain concerns, such as crime. When one considers the nature of the pandemic, fear of COVID-19 becomes less about particular and potentially controllable instances of threat. Therefore, the generalized social trust common of the religiously committed may not mitigate the fears of COVID-19, leading to a concern that may inspire a gun purchase. While this is a possible explanation, there are concerns about statistical power given the small sample size. More exploration is certainly needed.
Finally, our study does have limitations. One data limitation already mentioned is the relatively small sample size of gun owners and recent purchasers. Even though our survey sample was representative of American adults, a larger overall sample would have allowed us to better assess the motivations of those respondents that purchased firearms in 2020. In addition, a more systematic analysis could also include direct measures of Christian nationalism and more traditional religion measures. Other researchers should consider including these measures in future research examining the role of religion in gun purchasing.
We do think this project is an important initial step in understanding the gun purchase spike in America during the turbulent year of 2020. Many of the concerns aroused in 2020 will continue, including political polarization, the ongoing threat of COVID, and civil unrest over social justice issues. Social scientists should continue to consider how events are interpreted through religion and how these factors may shape future gun culture.
____________
Material referencing an appendix in this chapter can be found online available here: https://dataverse.harvard.edu/dataverse/epidemic_among_my_people.