First Monday

Crowdfunding during COVID-19: An international comparison of online fundraising by Greg Elmer, Sabrina Ward-Kimola, and Anthony Glyn Burton

This article performs a digital methods analysis on a sample of online crowdfunding campaigns seeking financial support for COVID related financial challenges. Building upon the crowdfunding literature this paper performs an international comparison of the goals of COVID related campaigns during the early spread of the pandemic. The paper seeks to determine the extent to which crowdfunding campaigns reflect current failures of governments to supress the COVID pandemic and support the financial challenges of families, communities and small businesses.


Off-line crowdfunding & the GoFundMe platform
Data collection and coding
International comparisons: COVID-19 crowdfunding around the world




With a few notable exceptions such as Germany, Taiwan, South Korea and New Zealand, governments around the world have come under heavy criticism for their poor handling of the global COVID-19 pandemic. At the outbreak of the pandemic questions largely focused on why pre-existing government plans, agencies, protocols and infrastructure were not adequately meeting the various demands posed by the viral outbreak. Many governments had been warned by their medical and health agencies after successive SARS and Ebola outbreaks that even more disruptive and deadly pandemics were inevitable [1]. In Canada, though, while the little known Global Public Health Intelligence Network (GPHIN) had been developed as a robust early warning system for emergent viral health threats (Oved, et al., 2020), the network seemingly failed to sufficiently inform or convince the Canadian government to enact policies that would have slowed the spread of the virus across the country. Countless news articles from around the globe told a similar story: government plans were inadequate, failing or being ignored by government decision-makers and political leaders.

As COVID-19 moved from an illness contracted while traveling overseas to a domestic virus spread within local communities (Miller, 2020) attention shifted to concerns over the capacity of local, regional and national health care systems, namely hospitals, and the availability of urgent medical supplies and personal protective equipment (PPE) for critical care staff. Stockpiles of PPE and other supplies around the world however proved to be wholly inadequate, even after governments were warned of the severity of the impending pandemic (Syal, 2020).

While many under-protected hospital staff have been infected or tragically lost their lives to COVID-19, the lack of government oversight, regulation and partial privatization of the long-term care sector has been blamed for the most deadly site of COVID-19 infections. By early May 2020 a staggering 82 percent all COVID-19 related deaths in Canada had been traced to long-term care homes (MacCharles, 2020). After years of government neglect and privatization the Toronto Star found that during the early outbreak of COVID-19 “For-profit nursing homes [had] four times as many COVID-19 deaths as city-run homes.” (Oved, et al., 2020) With a near collapse of the sector at hand, the Ontario government ordered the military to send staff to help support long-term care homes. Weeks later, Canadian Forces issued one of the most damning reports on long-term care homes in Canada’s most populous province [2]. The report was so disturbing that a news outlet felt it necessary to add the following text to accompany the report: “Graphic warning: Here is the full report from the Canadian Armed Forces. Details included in the report may be disturbing.” (Casey and McQuigge, 2020)

The closure of schools, universities and many small businesses particularly in the restaurant, retail and entertainment sectors, has likewise exposed some governments’ inability — or leader’s unwillingness — to provide income and economic stimulus support to the growing number of unemployed, precarious small businesses and hard-hit economic sectors. And where support has emerged, for example in the case of South Africa, the inadequateness of one-time emergency income replacement funds has only served to heighten awareness of the longer-term implications of income insecurity and inequality, chronic unemployment and systemic inequality (Dawson and Fouksman, 2020). Even before the virus hit South Africa hard the unemployment rate in the country was a staggering 29 percent (Bronkhorst, 2020). Estimates following the closure of vast parts of the economy now forecast the unemployment rate as high as 50 percent (Charles, 2020). By comparison, at the end of April 2020 30 million Americans had filed for unemployment, raising the unemployment rate from 4 to 15 percent. The numbers were similar in Canada, with 13 percent unemployment, 9 percent in the U.K., 8.4 percent in Italy, and 5.8 percent in Germany (Kretchmer, 2020).

While the scope of the crisis is unmatched in generations, none of these challenges to — and failures of — the welfare state are new. Many government programs for example were pushed to the limit, and later intensely critiqued after the financial shock produced by the 9/11 attacks, and then later after the 2008 global financial crisis. Moreover, with the decades-long forms of state austerity, privatization and deregulation, small businesses, communities, individuals and families have long known that that state cannot be trusted to support them during a time of need.

The acceleration of neo-liberal policies, leading to the downsizing of governments around the world, has witnessed the rise of self-help solutions, volunteerism and government programs to encourage entrepreneurship (Kotsko, 2018). The rise of the Internet, which disintermediated and disrupted established players and sectors of the economy from the music and television industry to banking and finance, has also served as a catalyst for raising start up and operating funds outside of the banking and government sectors (Giudici, et al., 2012). A recent report estimated that the total crowdfunding sector for example will increase by over $US124 billion over the next four years (starting 2020), a compounded annual growth rate of 18 percent (Global Newswire, 2020). Financial appeals to support entrepreneurial initiatives and technological innovations are now commonplace on crowdfunding sites like Kickstarter. Other crowdfunding sites like Indiegogo are designed to serve the film industry, a sector that has been devastated by cuts to independent producers. In this paper we focus on the GoFundMe crowdfunding platform, one that has traditionally served to host financial appeals from individuals and groups in financial trouble, often due to health-related challenges

Given the failures of governments noted above, our essay questions what recently launched campaigns on GoFundMe (between 27 March and 27 April 2020) can tell us about financial challenges faced by individuals, workplaces, communications and nation-states during the COVID-19 crisis. We are particularly interested in determining the intent of such campaigns, and in so doing identifying common areas of financial precarity. Through a digital-methods (Elmer, et al., 2012; Rogers, 2013; Marres, 2017) approach to studying Web-based communication and networking, we seek to determine the financial goals of GoFundMe-launched crowdfunding campaigns. Digital methods focuses primarily on how the affordances of social media and the Internet produce possibilities and constraints for online practices and forms of organizing. In short, the paper asks, what are the financial appeals seeking to remediate or fund? In support of what issues or challenges? Through an analysis of 500 campaigns launched around the world, this essay investigates the initiators of such campaigns (individuals, community groups or organizations?), and national origin or location of the campaigns. In other words, this investigation frames the question of campaigners identity in relationship to communities and workplaces as opposed to demographics, helping to further qualify the goal of the crowdfunding campaign.

More broadly we ask, what can online fundraising, or ‘crowdfunding’, tell us about the current social anxieties and economic precarities (and the coping strategies) exacerbated by the COVID-19 pandemic? And lastly, we compare the answers to these questions across countries in an effort to determine the contrasting challenges posed by COVID-19 in each country. What gaps in and failures of government are exposed by such appeals? While our sample is relatively small (500), it nevertheless produced distinct findings and sharp differences between crowdfunding/financial appeals around the world. Such findings, furthermore, shed light on the growing fundraising sector during a time when government resources and political leadership are being seriously challenged by COVID-19, and severely constrained by the compounded impact of austerity cuts and neo-liberal policies.



Off-line crowdfunding & the GoFundMe platform

Crowdfunding has long established itself as a remedial practice established through various regimes of governmentality, and has most recently articulated through online iterations of decentralized fundraising networks. Crowdfunding-like practices long precedes the Internet; in lieu of the platform, the dynamic of crowdfunding holds in the collection plate, offertory, newsletter or flyer, all media forms that have proven conducive to financial contributions. Echoing the operative logic of GoFundMe, these early off-line fundraising endeavours concerned the welfare of those in need, particularly in lieu of suitable government support and response in times of health crises.

Prior to the development of the modern nation state, religious institutions were central in collecting resources and facilitating crisis response, and this dynamic was prescient in pandemic scenarios of the middle ages [3]. Financed largely through tithes, Confraternite [Christian voluntary organizations] of Renaissance Italy began as informal stand-ins where the government fell short, eventually developing into quasi-government bodies providing generalized respite to the urban poor [4]. Beginning in the late fifteenth century, Venerabile Arciconfraternita della Misericordia di Firenze offered plague-related services such as transporting those infected to town hospitals and burying the dead, to which they were met with increasing donations from the public to ensure continual provisions [5]. As the century turned, funds from various sources grew to provide quarantine housing in the form of wooden huts, basic necessities, medicine and professional health practitioners for their services [6]. Having developed into a ritualized, quasi-governmental alternative, Confraternite reflected early threads of contemporary austerity dynamics; a quick provision of services through distributed funding can be traced to public crises in the modern era — the 1918 Spanish flu and the HIV/AIDS epidemic of the 1980s and 1990s.

Just as Venerabile Arciconfraternita della Misericordia di Firenze acted as ‘first responders’ to Florence’s black plague in the late fifteenth and early sixteenth centuries, the Spanish flu and HIV/AIDS epidemics were met with considerable local action in lieu of government support. The mobilization of services and volunteers was marked in the United States during the 1918 Spanish flu, as the majority of government resources were being channeled to the war effort overseas [7]. As a result, charged with the role of public health response, the severely under equipped American Red Cross relied upon the resources and coordination of localized groups throughout the country [8]. Likely due to economic hardship within a wartime economy, response took the form of services, lodging and soup kitchens in lieu of cash contributions to a centralized public health agency [9]. It must be noted that within these efforts, Black Americans were met with discriminatory policies, which led to a multiplicity of informal yet segregated community responses [10]. This prevailing theme of community-centered approach is later reflected at the outset of the HIV/AIDS crisis in the United States beginning in 1981. Local activist groups — ACT UP in Los Angeles and Gay Men’s Health Crisis in New York — were among the first groups fundraising and organizing informal relief programs for those affected [11]. Reflecting the trend of the previous two case studies, these localized bodies were borne from government inadequacy; research pointed to Reagan’s allegiances with the religious right for not having allocated research funding until 1983, or publicly addressing the epidemic until 1985 [12]. Originating from disparate contexts, these case studies exemplify the societal tendency to organize alternative financing schemes in times of crisis. Further, if the explicit exclusion of Black and LGBTQ2+ Americans serves as any indication, alternative schemes are disproportionately necessary within intersectional or lower-income communities due to waning and/or discriminatory social spending policies.

Outlining an earlier history of crowdfunding dynamics helps to orient the newer developments unfolding in concert with the emergence of the GoFundMe platform. The year 2008 marks the coalescing of Web 2.0 infrastructures and the American financial crisis, leading to the emergence of Internet crowdfunding in response to the crippling of medical and social assistance programs (Berliner and Kenworthy, 2017). As GoFundMe grew, fundraisers were obligated to market themselves, appealing to the emotions of potential donors with well-composed stories of personal crises (Berliner and Kenworthy, 2017). Berliner and Kenworthy note that nearly a fifth of GoFundMe’s campaigns seek respite from medical bills worldwide, indicating a major market in alternative healthcare fundraising, wherein campaigners craft tearjerker descriptions to stand out from the crowd. This style of crowdfunding has been described in the scholarly literature as affective in contrast to reward based (Gehring, 2016). The reward-based era of early 2010’s crowdfunding set the stage for the subsequent affective era of campaigning seen up to the present. Heavily influenced by Silicon Valley entrepreneurialism that preceded contemporary business to consumer (B2C) enterprise models; reward-based campaigns provided the tools that entrepreneurs could use to elicit funds for new consumer prototypes and services, offering perks such as tee-shirts to financial contributors and supporters. As online crowdfunding increased in popularity, so too did the need to develop individual financial appeals and master the technology required to fundraise for precarious individuals and communities. Lilly Irani (2015) refers to this phenomenon as “entrepreneurial citizenship”, where technological innovation strives to increase “the size of the economic pie as an alternative to redistributing it” and where the subject-entrepreneur collaborates with a wider collective rather than a self-interested agent. In this view, the crowdfunding entrepreneur is not only striving for individual success, but the success of their wider communities in lieu of unreliable, state-funded safety nets.

As the go-to fundraising platform during financial crises, GoFundMe has witnessed a dramatic rise in appeals for financial help during the outbreak of the COVID-19 pandemic. Soon after the World Health Organization (2020) declared a global pandemic 35,000 new COVID-related crowdfunding campaigns were launched on GoFundMe. This rapid expansion represents an approximately 60 percent jump in campaigns hosted by the platform in mid-March 2020, appeals that must compete for attention among a sea of upsetting personal stories of financial need. As a result, the entrepreneurial spirit of GoFundMe has exacerbated as COVID-19 plunges more people into precarity, with widespread calls for help necessarily being supplemented with what amounts to small-scale marketing campaigns. While we have gone to some lengths to link the history of government deregulation and privatization to the rise of affective and entrepreneurial (reward-based crowdfunding), the COVID-19 pandemic offers some unique characteristics and dynamics that deserve in-depth, empirical study. First, while clearly producing an economic crisis for governments, communities, families, companies and individuals, the crisis is rooted in issues of community health. We would anticipate then that the hyper-individualism, promoted by neo-liberalism and entrepreneurial culture, might be tempered on GoFundMe campaigns given the viral, community implications and impact of the virus. In other words, the pandemic challenges the intensely individualistic approach to fundraising previously witnessed on platforms such as GoFundMe. COVID-19 is, moreover, not just a challenge to individual’s health, it has proven to be a deadly threat, particularly to immuno-compromised individuals, communities with poor health and elderly populations. Such populations are not among the first one might think of in relation to entrepreneurial appeals online or crowdfunding at all.

Before turning to our study, it is important to note that of course neo-liberal, austerity and privatization policies are not spread evenly — internationally or domestically, nor are they in Organization for Economic Development (OECD) countries. Such policies are also exacerbated by political instabilities and a lack of trust in government [13]. The uneven distribution of economic power, and subsequent ability to withstand public health and economic related challenges posed by COVID-19 also call forth important comparative methods to better understand the contrasting appeals launched by individuals and groups in different countries and regions.



Data collection and coding

As a study of online crowdfunding campaigns this study was highly dependent upon text and data encoded into GoFundMe hosted campaigns. GoFundMe’s lack of a public-facing application programming interface [14] (API), typically the gateway to collecting data on Internet-based platforms, posed unique challenges for this study. GoFundMe campaigns contain a rich set of metadata that can help researchers contextualize the ‘user-generated’ components of individual campaigns, such as the narrative pitch or appeal for financial contributions. This study therefore took some care to collect as much data as possible from the platform and its specific campaigns.

In order to collect data from the campaigns we circumvented GoFundMe’s lack of a public API by writing a scraper software program that collected hyperlinks for discrete campaigns [15]. Using the “covid” search query our software collected URLs for 1,430 GoFundMe campaigns. The scraper ran daily from 27 March to 27 April 2020, capturing campaigns published within this time frame and updating campaigns that had been previously captured. Our program then loaded the raw content for each campaign and parsed out key information including the campaign’s location, the campaign’s fundraising goal, the amount raised as of that date, the campaign’s category on the platform and user statistics such as the amount of donors, shares and campaign followers. To perform a qualitative coding of campaigns we next reduced this dataset of 1,430 campaigns down to a random sample of 500 campaigns by using the random data selection process facilitated by Python programming software.

Through an iterative, qualitative coding of our random sample of 500 GoFundMe “Covid” campaigns, we produced three parallel coding schemes (Table 1). These manual coding schemes proved more illustrative than the preset categories provided as options to campaigners by the platform [16]. A tertiary review of our campaigns found that campaigners often incorrectly labeled their campaigns using GoFundMe’s pre-set campaign categories, this due to both limited options or a poor choice by the campaigners.


Table 1: Coding schema.
Pandemic solutionPersonal protective equipmentCommunity members
Death/FuneralFriend/Family members
Community charity
Small businessThird party
Income insecurityFamily/FriendsThird party
Food/SuppliesAffected party
Global charity
Hospital fundraising


We first developed a primary coding scheme with two categories that divided campaigns into fiscal pathways, that is determining whether the financial need was immediate and urgent (‘income security’) or provided a more forward-looking, typically entrepreneurial campaign (‘pandemic solution’). This latter distinction might at first glance seem counterintuitive given GoFundMe’s role as a fundraising platform which promotes itself as remedial, however our data warranted an additional category reflective of a solutionist and entrepreneurial ethos of crowdfunding. Although this first code was insightful, we felt that it did not illustrate the multidimensional nature of the collected data. Consequently, we coded next for the indicated object warranting financial respite, in short what the collected money was going to purchase or remediate. In contrast to fiscal pathways, campaign objects can be difficult to discern due to vague or misleading campaign descriptions [17]. However, we developed this scheme iteratively through careful, qualitative analysis considering the campaign title, text and images, thus yielding 10 categories (Table 1). This secondary coding scheme illustrated spaces of everyday life that have been impacted, and the objects that receive the highest degree of funding. At this point, our coding schemes pointed to the why and what, however we did not have insight into who was initiating these crowdfunding campaigns. Our third coding scheme therefore indicates the actor involved in launching the campaign.

Before moving on to our findings, let us review the rationale behind these coding decisions. For the primary coding then, the funding pathways (future solutions or funds to alleviate lost income), indicates why the money is required; while the secondary coding, the object, indicates what the money is required for. Lastly, the tertiary coding identifies the relationship that the fundraiser has with the object of concern.




We begin our discussion of the findings by focusing on the entire dataset, determining the most prominent goals for all campaigns (secondary coding), the parties initiating fundraising campaigns (tertiary coding), and then lastly, the locations where the campaigns were launched. Upon starting a campaign, location is user generated via a zip or postal code. This geographic information will be used to compare GoFundMe campaigns across national contexts. After providing a cross-country analysis of our crowdfunding campaigns we turn to the crowdfunding hosts. The final section ends with a discussion of our primary coding, which helps to answer the larger question that this paper raises, specifically the diminishing role — or ability — of government to adequately address the range of issues posed by the COVID-19 pandemic.


Campaign objects (purpose of funds)
Figure 1: Campaign objects (purpose of funds).


Our secondary coding schedule identified the fundraising object, the primary financial rationale or pitch for the crowdfunding campaign (Figure 1). Overall, we found that personal protective equipment or PPE was the most frequently identified issue for crowdfunding contributions (17 percent), followed closely by small business (15.6 percent), community charity (13.8 percent), food/supplies (10.8 percent) and family/friends (9.6 percent). The remaining and less represented codes — global charity, death/funeral, hospital fundraising, arts and sports and animal welfare each stood at eight percent or less of the overall share. Just over eight percent of the campaigns were in a language other than English, highlighting the linguistic and national bias of the platform.

While there was a strong cross section of financial appeals, highlighting the breadth of issues raised by the COVID-19 pandemic, GoFundMe campaigns under study focused overwhelmingly on matters of business, family, health and community. Five of such categories constituted 72 percent of all our campaigns, indicating that financial need was most urgent in PPE, small business, community charity, food/supplies and friends/family needs. This finding highlights potential or actual governmental gaps in the immediate needs of families and communities in everyday, local settings. Likewise, while PPE shortages have been a common international issue, campaigns seeking to produce, fund and distribute PPE largely focused on the bandwidth of local networks, for local health care workers and health providers. This local focus is also reflected in the output of community charity and food/supply campaigns, which address a wider set of needs, often for a specific socioeconomic or geographically located community.


Campaign actor (initiator of campaign)
Figure 2: Campaign actor (initiator of campaign).


Complicating this consideration of the local, we found that the most represented actor category for all campaigns to be a third party (32 percent), followed closely by a community member (31 percent) (Figure 2). This high representation of third parties relates to a combination of extensive third-party fundraising throughout all coded categories, as well as an overrepresentation of a third party in the five least represented campaign objects represented by the second coding schedule, particularly global charity, animals and arts/sports. If we focus in on the top five object categories, however, the relationship to community actors becomes more pronounced. Forty-eight percent of PPE-oriented campaigns, 54 percent of community charity, and 61 percent of food/supplies coded campaigns were initiated by self-identified community members. The remaining actor categories, affected party (21 percent) and friends/family member (16 percent), were relatively marginal and specific to small businesses, friends and family support, as well as funds elicited for death or funeral expenses.

The predominance of campaigns initiated by a third party also suggested either entrepreneurial and compassionate motivations on the part of individuals or small groups indirectly impacted by COVID-19. Figure 3 further highlights how indirectly affected or third party campaigners initiated a diversity of campaigns. While PPE served as the most predominant object of third-party-driven campaigns, over a quarter (27 percent) of the third party campaigns sought funds for international or global causes and charities. By comparison Figure 4 shows that while PPE also predominated the community member-initiated campaigns, Food/supplies were also a source of major concern, once again highlighting how everyday issues of income precarity were heightened by the COVID-19 pandemic.


Third party campaigns
Figure 3: Third party campaigns.



Community member campaigns
Figure 4: Community member campaigns.


Compared to third party and community member-initiated campaigns, affected party campaigns demonstrated the least diversified campaigns. Figure 5 below shows that a full 70 percent of crowdfunding campaigns were in support of small businesses. Many of these appeals contained urgent language suggesting that a failed campaign would lead to the closure of a business. One such campaign on behalf of East London’s Rio cinema describes the financial need as imminent, so as to “ensure that the COVID-19 pandemic does not jeopardize the future of this much loved Hackney institution” [18]. The other significant cause for concern and fundraising for affected parties was sadly death/funeral campaigns (13 percent), which we will see later in the national comparison section later was largely concentrated in U.S. based campaigns. Friend/family initiated campaigns (Figure 6) saw the largest number of death/funeral related appeals for funds (35 percent), along with general appeals to support friends and family who found themselves financially in a precarious position because of the pandemic (63 percent). Comparing Figures 5 and 6 and in particular Death/Funeral campaigns, we can also conclude that GoFundMe campaigns are more likely to be launched for the assistance of others, as a form of charity. Affected parties were much less likely to launch campaigns to support their own financial needs, at least with regards to Death/Funeral expenses.


Community member campaigns
Figure 5: Community member campaigns.



Community member campaigns
Figure 6: Community member campaigns.




International comparisons: COVID-19 crowdfunding around the world

Now that we have explored what these campaigns target and who they were launched by, we turn next to investigate where GoFundMe campaigns were being launched, and how the campaigns emphasized different appeals and issues across national contexts. A majority (67 percent) of crowdfunding campaigns in our sample originated within the United States, followed by England (14 percent), Canada (6 percent), and Italy (5 percent), all countries that faced significant COVID-19 infection rates. Nine other countries launched campaigns, representing just 6 percent of our sample. These findings are not particularly surprising given that the GoFundMe platform was launched and remains based in California. The platform also restricts language options to English (US), English (UK), Spanish, German, French, Italian, Dutch and Portuguese with a few notable exceptions (such as Brazil and Mexico) mirroring the countries represented in our sample.


Campaigns per country
Figure 7: Campaigns per country.


In terms of money collected, the U.S. based campaigns netted US$5,724,760 in total donations, US$1,069,592 in Italy, followed by US$578,311 in England, and US$162,579 in Canada. While more campaigns were obviously launched in the United States (331 vs. 65 in England, 29 in Canada, and 26 in Italy), Italy saw the highest average income per campaign at US$56,947 followed by Spain at US$49,643, and then the United States at US$17,295. This finding may point to the competition for dollars and donations in the larger U.S. marketplace, or the more focused and local appeals of campaigns in these other countries (Popper and Lorenz, 2020).


Campaign types
Figure 8: Campaign types.


While many countries launched a diverse set of campaigns, the U.S., Canada and England launched the most diverse set (Figure 8). PPE campaigns constituted the most launched campaigns in all countries except for the United States and Australia. This finding mirrors the widespread coverage that this issue raised around the world. Indeed, during the initial outbreak of the pandemic, access to PPE became a de facto site of competition between countries, provinces, states and municipalities (Chacko, 2020). This finding again supports the view that many countries were woefully unprepared to supply critical protective supplies for front line health workers. GoFundMe campaigns that focused on fundraising for PPE were common across our multinational sample. England saw the highest number of PPE fundraising, representing 28 percent of our total PPE category, and 35 percent of all campaigns within that country. These findings mirror reports around the world, for example the British Medical Association survey that found 48 percent of doctors in the National Health Service (NHS) either bought or received their masks from local charities (Press Association, 2020). A final finding of note is the high number of hospital fundraising observed in Italy, a marginal category in every other country. Italy, among one of the first hardest hit countries by the pandemic, observed a relatively disproportionate degree of hospital fundraising with a single campaign raising over US$400,000. Suffering from critically cramped hospitals and a high number of physician deaths, the disproportion likely accounts for the severe lack of medical support being ramped up in time to house the exponentially admitted patients (Reguly, 2020).

The United States by comparison saw a much larger percentage (21 percent) of small business driven campaigns compared to all other countries. Only 15 percent of all Canadian GoFundMe located campaigns focused on small business needs, while the number dropped even further in England (5 percent). The second most launched U.S. based campaign focused on community charities. Over 33 percent of campaign funds in the U.S. were donated to Community charity crowdfunding campaigns, compared with 22 percent in Canada. The high number of United States based community charity campaigns can in part be explained by the loss of service jobs, and also the precarious conditions of indigenous communities living on reservations, immigrants, undocumented workers, groups who have been particularly hard hit both in terms of economy and health as a result of the COVID pandemic. And when we view the amount of money collected (Figure 9), we find community charities to be even more significant in the United States. The highest grossing campaigns within our community charity coding category include relief funds for Navajo and Hopi families organized by the Rural Utah Project Education Fund [19], as well as assistance for immigrants and undocumented workers who are not-eligible for state or federal COVID-19 response benefits [20]. Although targeting different cohorts, these community oriented projects distribute aid and funds through third-party vetting processes in lieu of appropriate government assistance setups. This is reflected in the Canadian context, which observed a nearly equal split among food/supplies, community charity and PPE, all categories with concern to the local.


Funds received
Figure 9: Funds received.


These cross-country findings echo the heterogeneity of needs both locally and internationally, however an overrepresentation of certain objects suggests country-specific failures on behalf of prevailing support networks. Prevailing political debates within different jurisdictions may also explain these differences. For example, the cross-partisan consensus on the need to “protect” the National Health Service (NHS) [21] during the recent U.K. election may also explain the relatively high number of appeals for PPE in England-based crowdfunding campaigns. Differing levels of government support for unemployment and health insurance which are much more robust in England and Canada, compared to the United States, could also explain the different emphases and donations across national contexts. Our primary coding further highlights the crowdfunding approaches taken in the United States and in the other Western countries. While we might expect to see a larger percentage of entrepreneurial campaigns, as represented by “solution” based campaigns that offered future solutions, the precarious nature of income security in the U.S. even at this relatively early stage of the COVID-19 pandemic demonstrates that half of the campaigns launched in the US sought funds for urgent and immediate financial help.


Campaigns as solutions or income insecurity
Figure 10: Campaigns as solutions or income insecurity.





This paper began by contextualizing the growth of crowdfunding online as an entrepreneurial attempt to develop revenue streams for new innovative solutions. Part of this new wave of online fundraising however also included urgent and often affective appeals for help brought on by devastating personal circumstances and crises. We might at first glance posit these two approaches to crowdfunding as unique and distinct, yet both are emblematic of a withering of state support for many different sectors in society, including of course welfare, unemployment support and public universal health benefits and funding for the health sector writ large.

As a relatively small case study of crowdfunding campaigning at the early outbreak of the COVID-19 pandemic, particularly in the United States, this study has its limitations. Most important, back to our guiding question, can we fully and convincingly point to the growth of GoFundMe campaigns as evidence failed neo-liberal policies and austerity programs? Our history section clearly noted that fundraising, particularly for health-related issues, was often initiated by third parties — and we found a similar dynamic in contemporary online fundraising campaigns. While crowdfunding mirrors many aspects of neo-liberal governmentality, we can also conclude that these crowdfunding campaigns also provide insight into affective responses to the pandemic, that is highlighting both the social anxieties and economic precarities produced by the COVID-19 pandemic. This was supported by the finding that many unaffected individuals and groups launched campaigns. Individual third parties felt the need to do something to help. In other words, the study found that GoFundMe covid campaigns are often altruistic endeavours, they are not simply launched by those directly in need. Hence, such campaigns are launched by those imagining in some instances we can say, the greatest and most obvious needs in their own surroundings and media space. The importance of community fundraising therein repeated itself throughout the various approaches we brought to our crowdfunding sample. While there were many international issues and goals, the vast majority of the campaigns spoke to matters of local concern, concerns that were already prevalent and latent. National concerns were most prevalent in the U.S. and England based campaigns, both connected in different ways to the state of health care in those respective countries. England and the rest of the U.K. continues to respond to robust political campaigns in support of their national health service, while American health insurance and coverage continues to be a deeply partisan issue that continues to divide American political opinion.

As the crisis spreads to other countries around the globe and intensifies in the United States, Brazil, Spain and other nations, further studies are needed to track the national and regional responses to the pandemic. Unlike other Internet-based platforms such as Twitter and Facebook, GoFundMe and other like-minded crowdfunding sites offer a distinct worldview, a user-generated view of urgent financial appeals and campaigns to care for communities suffering from economic deprivation and government cuts. Expanded studies of crowdfunding would offer important stories and lessons for policy-makers and community leaders in their efforts to prepare for subsequent waves of the pandemic and in the long term, debates over how best to reconstruct our communities, health systems and economics once the crisis has subsided. End of article


About the authors

Greg Elmer is is Bell Globemedia Research Chair and Director of the Infoscape Research Lab at Ryerson University in Toronto, Canada.
E-mail: gelmer [at] ryerson [dot] ca

Sabrina Ward-Kimola is a M.A. student in the Graduate Program in Communication and Culture at York & Ryerson Universities.
E-mail: swardkimola [at] ryerson [dot] ca

Anthony Glyn Burton is a Ph.D. student in communication at Simon Fraser University.
E-mail: Anthony_burton [at] sfu [dot] ca



1. Cf., in the case of the United States:


3. Manetti, et al., 2017, p. 511.

4. Ibid.

5. Manetti, et al., 2017, p. 516.

6. Manetti, et al., 2017, p. 517.

7. Jones, 2010, p. 101.

8. Jones, 2010, p. 102.

9. Ibid.

10. Gamble, 2010, p. 119.

11. Gould, 2009, p. 52; Roth, 2017, p. 2.

12. Gould, 2009, p. 49.

13. In the American case cf.,

14. An application program interface consists of a set of functions and procedures that allows for the creation of applications that access the data of a Web page, in our case.

15. This software is available for public use, at Gofindme was written for this project. GoFundMe does not have a public-facing API for easy access to campaign data; this program scripts a call to the MVC API that GoFundMe uses to populate its user-facing search page with results. The program takes a given search term and a number of iterations to call this search and makes a formatted call to this API to return the search results programatically without the use of a Web browser. This allowed us to capture information about campaigns that is shown on GoFundMe’s search results page: the campaign name, the location of the campaign, its full description, the amount that the campaign has raised at the time of the script, the amount that the campaign runners are asking for and a link to the full campaign page. The program then places each result into a SQLite tabular database.

To capture a more robust dataset for each campaign, the program takes all campaign URLs captured and grabs information from each full campaign page. Since the campaign pages themselves are not accessible through the same MVC interface as the search results, the program loads each page in an automated browser using the Selenium Web driver. It then scrapes from the page all initial information from the search results card, plus information not displayed in the search results: the category that users place their campaign into, the amount of donors, the amount of shares and the amount of followers of a campaign.

The program has one key limitation since it uses scraping as opposed to a public-facing API. For each search query, a maximum of 113 pages of results will load. Given nine cards displayed per search page, each iteration of the search returns a maximum of 1,017 results despite GoFundMe often displaying the number of results for “covid” in the range of 60,000 to 90,000. We attempted to mitigate this by running the search query five times for every run and running the search every day. This allowed us to acquire slightly different sets of 113 pages from which we were able to capture more than the initial 1,017. This daily search also allowed us to track the date on which campaigns were deleted or removed. By instructing the Web driver to run through the complete and growing set of URLs recorded in the database, the date that a previously captured URL returns a “campaign does not exist” error was recorded.

16. The platform allows campaigners to select a thematic category for their campaign. Pre-set options for campaign further support our characterization and branding of the GoFundMe campaign as focused on remedial personal and health related crises. The platform’s categories include the following: Accidents & Emergencies; Medical, Illness & Healing; Volunteer & Service; Community & Neighbors; Animals & Pets; Business & Entrepreneurs; Funerals & Memorials; Creative Arts, Music & Film; Sports, Teams & Clubs; Travel & Adventure; Education & Learning; and Babies, Kids & Family.

17. We found no clear examples of fraudulent or propagandistic campaigns in our sample.

18. GoFundMe, “Support the Rio Cinema during the COVID-19 crisis,” at, accessed 27 April 2020.

19. GoFundMe, “Navajo & Hopi Families COVID-19 Relief Fund,” at, accessed 27 April 2020.

20. GoFundMe, “MN Immigrant Families COVID-19 Fund,” at, accessed 27 April 2020.

21. Protecting the NHS became a rally cry across the U.K. over the past few years; cf., Similar language was evidenced in many PPE focused crowdfunding campaigns in the U.K., one of which noted the need for example to “protect front-line health care workers” in the NHS; see



Lauren S. Berliner and Nora J. Kenworthy, 2017. “Producing a worthy illness: Personal crowdfunding amidst financial crisis,” Social Science & Medicine, volume 187, pp. 233–242.
doi:, accessed 15 October 2020.

Quinton Bronkhorst, 2020. “South Africa’s unemployment rate unchanged at 29.1%,” BusinessTech (11 February), at, accessed 15 October 2020.

Liam Casey and Michelle McQuigge. “‘Disturbing’ Ont. long-term care home report doesn’t come as a surprise to families,” CTV News (27 May), at, accessed 15 October 2020.

Diya Chacko, 2020. “Coronavirus today: The competition for PPE,” Los Angeles Times (10 June), at, accessed 25 June 2020.

Marvin Charles, 2020. “South Africa’s unemployment rate expected to reach 50%,” IOL (27 May), at, accessed 15 October 2020.

Hannah J. Dawson and Elizaveta Fouksman, 2020. “Why South Africa needs to ensure income security beyond the pandemic,” The Conversation (20 April), at, accessed 15 October 2020.

Greg Elmer, Ganaele Langlois, Fenwick McKelvey, 2012. The permanent campaign: New media, new politics. New York: Peter Lang.

Vanessa Northington Gamble, 2010. “‘There wasn’t a lot of comforts in those days:” African Americans, public health, and the 1918 influenza epidemic,” Public Health Reports, volume 125, supplement 3, pp. 113–122.
doi:, accessed 15 October 2020.

David Gehring, 2016. “The depoliticized politics of crowdfunding: A critical examination of the Darren Wilson crowdfunding campaign,” First Monday, volume 21, number 10, at, accessed 15 October 2020.
doi:, accessed 15 October 2020.

Giancarlo Giudici, Ricardo Nava, Cristina Rossi Lamastra and Chiara Verecondo, 2012. “Crowdfunding: The new frontier for financing entrepreneurship?” SSRN (6 October).
doi:, accessed 15 October 2020.

Global Newswire. 2020. “Global crowdfunding market (2020 to 2024) — Players include Kickstarter PBC, Patreon & Teespring among others” (10 June), at, accessed 3 June 2020.

Deborah B. Gould, 2009. Moving politics: Emotion and ACT UP’s fight against AIDS. Chicago: University of Chicago Press.

Lilly Irani, 2015. “Hackathons and the making of entrepreneurial citizenship,” Science, Technology, & Human Values, volume 40, number 5, pp. 799–824.
doi:, accessed 15 October 2020.

Marian Moser Jones, 2010. “The American Red Cross and local response to the 1918 influenza pandemic: A four-city case study,” Public Health Reports, volume 125, supplement 3, pp. 92–104.
doi:, accessed 15 October 2020.

Adam Kotsko, 2018. Neoliberalism’s demons: On the political theology of late capital. Stanford, Calif.: Stanford University Press.

Harry Kretchmer, 2020. “How coronavirus has hit employment in G7 economies,” World Economic Forum (13 May), at, accessed 15 October 2020.

Tonda MacCharles, 2020. “82% of Canada’s COVID-19 deaths have been in long-term care, new data reveals,” Toronto Star (7 May), at, accessed 15 June 2020.

Giacomo Manetti, Marco Bellucci and Luca Bagnoli, 2017. “The management of the plague in Florence at the beginning of the Renaissance: The role of the partnership between the Republic and the Confraternita of Misericordia,” Accounting History, volume 22, number 4, pp. 510–529.
doi:, accessed 15 October 2020.

Noortje Marres, 2017. Digital sociology: The reinvention of social research. London: Polity.

Adam Miller, 2020. “‘The time is now to act’: COVID-19 spreading in Canada with no known link to travel, previous cases,” CBC News (16 March), at, accessed 3 June 2020.

Marco Chown Oved, Brendan Kennedy, Kenyon Wallace, Ed Tubb and Andrew Bailey, 2020. “For-profit nursing homes have four times as many COVID-19 deaths as city-run homes, Star analysis finds,” Toronto Star (8 May), at, accessed 29 May 2020.

Nathaniel Popper and Talor Lorenz, 2020. “GoFundMe confronts coronavirus demand,” New York Times (26 March), at, accessed 1 June 2020.

Press Association, 2020. “Nearly half of England’s doctors forced to find their own PPE,” Guardian (3 May), at, accessed 25 May 2020.

Eric Reguly, 2020. “Italian doctors’ fatalities reach tragic levels as they fight COVID-19 in overburdened hospitals,” Globe and Mail (3 April), at, accessed 20 May 2020.

Richard Rogers, 2013. Digital methods. Cambridge, Mass.: MIT Press.

Benita Roth, 2017. The life and death of ACT UP/LA: Anti-AIDS activism in Los Angeles from the 1980s to the 2000s. Cambridge: Cambridge University Press.
doi:, accessed 15 October 2020.

Rajeev Syal, 2020. “Government ignored warning to stockpile PPE as COVID-19 spread,” Guardian (11 June), at, accessed 1 June 2020.

World Health Organization, 2020. “Archived: WHO timeline — COVID-19” (27 April), at, accessed 27 May 2020.


Editorial history

Received 29 June 2020; revised 7 September 2020; revised 8 September 2020; accepted 8 September 2020.

Copyright © 2020, Greg Elmer, Sabrina Ward-Kimola, and Anthony Glyn Burton. All Rights Reserved.

Crowdfunding during COVID-19: An international comparison of online fundraising
by Greg Elmer, Sabrina Ward-Kimola, and Anthony Glyn Burton.
First Monday, Volume 25, Number 11 - 2 November 2020