The nonexcludable function of sports stadiums in climate-changed cities

Abstract

The unrelenting threat of climate change and extreme weather events on human settlements has prompted urban planners and public policymakers to develop robust climate mitigation and adaptation strategies for their cities. As part of these measures, they may identify existing or proposed infrastructure that could serve residents during a public emergency (e.g. as shelters). One such candidate is the modern professional sports stadium, which has demonstrated its versatility over the past several years by serving as field hospitals, COVID-19 testing and mass vaccination sites, and municipal voting centers. In this commentary, we consider the essential contribution sports stadiums may make to cities and regions impacted by extreme weather and climate change. Specifically, we examine the utility of a publicly funded stadium as a shelter in response to large-scale disasters. As part of our analysis, we identify the structural, geographical, organizational, and logistical factors that impact the feasibility of this proposal.

Keywords:

• sport ecology
• sustainability
• climate change
• public sport facilities
• public subsidies
• political science

Introduction

In 2020, amid record-breaking wildfire and hurricane seasons, Americans spent more nights in emergency shelters than in any year of the previous decade (Flavelle 2020). Anthropogenic climate change and the consequential threat of extreme weather events to human settlements have motivated some urban planners and public policymakers to implement new measures of climate mitigation and adaptation in their cities (Aylett 2014; Dhar and Khirfan 2017). Considering urban areas account for nearly half of the world’s population (United Nations 2019), urban planners and policy experts have argued that adaptation efforts at the city scale carry significant potential to meaningfully increase emergency readiness for the public (e.g. Elvas et al. 2021; Hallegatte and Corfee-Morlot 2011). Part of a city’s climate plan may include enhancing its temporary housing plan, upgrading its permanent shelters, and—central to this article—identifying existing or proposed infrastructure that could serve residents during a public emergency (McConnell and Bertolin 2019). In this article, we argue that the modern sports stadium,¹ with its efficient and hardy design, large capacity and adaptable floorplan, and high-volume transportation and parking capability, is a particularly appealing facility to deliver public services and provide refuge to a community in times of emergency.

There is precedent for using sports venues as emergency response centres (ERCs; e.g. Blumenthal 2005; Brady 2007). During World War II, Rodney Parade in Newport, Wales, was used as an air-raid shelter for locals (Birt 2019), while the Saltdean Lido—a public pool in Brighton, England—was taken over by the National Fire Service to serve as a water tank for filling their trucks (Saltdean Lido n.d.). Meanwhile in London, Twickenham Stadium, the ‘home of rugby’, was used as a civil defence depot and earmarked as a decontamination site in the case of a chemical attack on London (Twickenham, 2019). Since then, the use of sports facilities as ERCs has proliferated globally, though typically in a makeshift, last-minute, responsive manner, rather than as a planned adaption measure. In 2020 alone, American stadiums were transformed into field hospitals, COVID-19 testing sites, and municipal voting centres (Orr and Kellison 2020). By 2021, sports facilities were also widely used as mass vaccination sites (Sears, Cianfrone, and Kellison 2021). When used in such a public-serving capacity, the sports stadium may be firmly established as a public good.

That professional sports stadiums are public goods—centres of civic pride, community identity, and image building that can be enjoyed infinitely and without exclusion (Johnson et al. 2012)—is a common refrain among stadium boosters, particularly as they advocate for public financing to construct or renovate a venue. On the other hand, anti-subsidy groups have challenged the nonexcludable component of the public-good argument, contending that the few benefits stadiums do afford to the community are vastly outweighed by public harms like gentrification (McGehee et al. 2018). Furthermore, opponents argue, any benefit is primarily produced by the teams and events within the stadium, and to obtain these benefits, residents must have the financial means to access the stadium (e.g. by purchasing tickets; Ahmed 2009). By definition, this practice would exclude members of the public from the stadium’s supposed benefits.

Assuming there is a desire by policymakers to maximize a professional sports stadium’s status as a public good, one strategy to reinforce its nonexcludability is to expand the services it provides to the community at large. In this paper, we consider the essential contribution sports stadiums may make to cities and regions impacted by extreme weather and climate change. Specifically, we examine the utility of a publicly funded stadium as a shelter in response to large-scale disasters. A response plan that proactively designated a city’s stadium as an emergency response centre (ERC; e.g. a temporary congregate mega-shelter, field hospital, command centre, or food and supply distribution centre), would not only address the problematic ‘management-by-disaster mentality’ (Feltmate as quoted in Bozikovic 2015, para. 7) that is prominent throughout U.S. cities, but it could also underscore the stadium as a public asset, thereby justifying a portion of public investment in the venue.

Climate change in cities

Several studies have identified the potential impacts (and benefits) of climate change on cities (e.g. Hunt and Watkiss 2011; Klein et al. 2007). Among the most damaging impacts, scientists have identified sea level rise and storm surges causing infrastructural damage and flooding; extreme events damaging built infrastructure (e.g. hurricanes or wildfires damaging buildings and roadways); health issues linked to higher average temperatures or extreme events (e.g. heat-related illness, extreme cold events causing hypothermia); changing patterns of energy use (e.g. putting higher demand on heating and cooling); and shifting water availability (IPCC 2022). Among these effects, storms are currently the costliest weather events in the developed world. According to the National Oceanic and Atmospheric Administration (2021), ‘of the 258 U.S. weather disasters since 1980, tropical cyclones [hurricanes] have caused the most damage: $945.9 billion total, with an average cost of almost $21.5 billion per event.’ Moreover, the Association of British Insurers (ABI; 2005) estimated that by the 2080s, climate change could cause a 75% increase in costs of damage in a hurricane season in the southern US, a 65% increase in costs of insured damage in a hurricane season in Japan, and a 5% increase in wind-related insured losses from extreme European storms. Already in the years since the ABI study, their predictions are on track: both 2011 and 2017 saw record-setting hurricane seasons with 16 events across the southern US, and in 2020, the record was beaten by 22 events (National Oceanic and Atmospheric Administration 2021).

Climate change is experienced differently across cities and regions, as it may manifest in several ways, including water stress, extreme rainfall, wildfire, sea level rise, heat stress, and hurricanes (Thompson and Serkez 2020). In addition to the extent and type of risk they face, cities may also vary in their preparedness for climate-related disasters. One tool to illustrate these differences is the Notre Dame Global Adaptation Initiative’s (2018) Urban Adaptation Assessment (UAA), which measures a city’s climate risk (i.e. vulnerability to climate change based on exposure, sensitivity, and adaptive capacity) and readiness (i.e. ‘capacity of an urban society…to mobilize adaptation investments from private sectors, and to target investments more effectively’; 3). Based on the UAA, we analysed 50 American cities with at least one major men’s professional sports stadium (i.e. MLB, NBA, NFL, NHL). In Table 1, we list ten major-league cities where risk scores exceeded readiness scores most, ranked by the size of the gap between scores. The table also reports each city’s primary climate threat(s) using data from Four Twenty Seven (a firm that predicts climate risk using climate models and historical records and mapped by Thompson and Serkez 2020).

Table 1. Major-league cities with largest climate risk–readiness differences.

Rank	City	Population	Primary Climate Hazard(s)	UAA Score	Metro Stadiums
1	Inglewood, CA	376,933	Water Stress	43.42	NFL
2	Anaheim, CA	349,964	Water Stress	31.56	MLB, NHL
3	Miami, FL	987,620	Heat, Hurricanes, Water Stress	27.04	MLB, NBA
4	Newark, NJ	281,054	Hurricanes, Water Stress	24.5	NHL
5	Los Angeles, CA	2,603,620	Water Stress	17.39	MLB, NBA/NHL, NFL
6	Buffalo, NY	254,480	Flood	16.76	NFL, NHL
7	Chicago, IL	2,709,534	Flood, Heat, Water Stress	15.43	MLB (2), NBA/NHL, NFL
8	Sacramento, CA	1,157,116	Flood, Heat, Water Stress, Wildfire	14.5	NBA
9	New York City, NY	8,419,316	Hurricanes, Sea Level Rise, Water Stress	12.18	MLB (2), NBA, NBA/NHL, NFL
10	Detroit, MI	674,841	Flood	8.79	MLB, NBA/NHL, NFL

UAA Score is difference between city’s Risk and Readiness scores (Notre Dame Global Adaptation Initiative 2019). List of primary climate hazards from Thompson and Serkez (2020). Metro Stadiums are count of professional sports venues within 12 miles of a city centre or county centre of population (Kellison 2021; U.S. Census Bureau 2019b). Population numbers from U.S. Census Bureau (2019a).

The city scale is increasingly being recognized for mitigation action. A principal benefit of grounding climate change adaptation efforts at the city scale is that it may make the associated risks more relevant to the private and public agents charged with designing and implementing possible responses (Albers et al. 2015; Chu, Anguelovski, and Roberts 2017). For example, analysis at the city scale is likely to coincide more closely with local administrative boundaries, thereby facilitating decisions related to adaptation at an appropriate level of governance. Indeed, the Intergovernmental Panel on Climate Change (IPCC) has concluded with ‘very high confidence’ about the extent to which industry, settlements, and society are capable of considerable adaptation in response to climate change (Klein et al. 2007). The IPCC’s confidence in adaptive capacity depends largely on the competence and capacity of individuals, communities, enterprises, and local governments, and their access to financial and other resources (Hunt and Watkiss 2011).

Scholars in urban planning and policy have advanced the concept of civic capacity to explain how cities and communities can adapt to the impacts of climate change (Sarzynski 2015). Civic capacity is defined by Briggs (2008) as the ‘extent to which the sectors that make up a community are (1) capable of collective action on public problems (the resource dimension) … and (2) choose to apply such capabilities’ (13). Originally conceived in the context of urban education, civic capacity has been expanded to other forms of collective action in communities (Sarzynski 2015). It requires both the institutional resources to solve a problem and the willingness to act collaboratively within and outside government systems to solve it (Stone 2001). Such capacity develops over the long-term (typically decades) and is built on perceptions of benefits and costs of civic mobilization and the actual outcomes from prior collective problem solving (Briggs 2008; Sarzynski 2015). To be most effective, civic capacity must be built proactively, and a shared understanding of the problem must be reached.

In 2020, emergency expenses related to natural disasters totalled $95 billion (Smith 2021), made worse by climate change and compounded by humanitarian crises like the COVID-19 pandemic. In the US, like elsewhere, emergency response funding is typically divided between the federal and state levels and, to a lesser extent, local government and the private sector (Pew Charitable Trusts 2020). The distribution of physical resources, activation and maintenance of shelters and ERCs, and implementation of recovery plans typically occur at the local level in partnership with higher levels of government. As disasters worsen, there is a need for greater collaboration between governments and a reassessment of available resources (financial, infrastructural, and social) to build the civic capacity that will reduce emergency costs moving forward and hasten the recovery time.

The cities identified in Table 1 (and indeed, others throughout the US) are not fully prepared to confront the inevitability of extreme heat or cold, rising seas, droughts, and other climate hazards. Efforts to improve local climate preparedness must include strategies to temper the effects of extreme events and the infrastructure required to support residents when an emergency occurs. In the following section, we discuss how sports stadiums can serve this purpose.

Stadiums as public gathering sites

Stadiums as public goods, and related arguments

In early 2021, National Football League commissioner Roger Goodell sent a letter to recently inaugurated President Joe Biden to express the league’s support of the nationwide COVID-19 vaccination effort. Specifically, Goodell committed to opening each NFL club’s stadium as a vaccination site:

The NFL and our 32 member clubs are committed to doing our part to ensure that vaccines are as widely accessible in our communities as possible. To that end, each NFL team will make its stadium available for mass vaccinations of the general public in coordination with local, state, and federal health officials. This is currently being done at seven NFL stadiums today. We can expand our efforts to stadiums across the nation more effectively because many of our clubs have offered their facilities previously as COVID testing centers as well as election sites over the past several months. (Goodell 2021)

Though brief, the letter underscored at least two interrelated points about modern stadiums. First, their size, location, and accessibility make them suitable settings for large-scale community events (e.g. voting centers, COVID testing and vaccination sites). Second, such use by the public may first require authorization from the stadium operator, ordinarily the team functioning as the venue’s primary tenant. The latter point is especially important considering the fact many U.S. stadiums are publicly owned but privately operated; that is, even if a city or county owns a stadium, it may not have the right to commandeer the facility for public use (e.g. Georgia World Congress Center Authority 2012; Washington State Major League Baseball Stadium Public Facilities District 2018).

Sports stadiums are not reliable engines of significant economic growth (Humphreys 2019; Matheson 2019b). Nevertheless, arguments that suggest otherwise can be convincing, particularly in cities debating whether it should subsidize a project to attract (or retain) a professional team or event (Connolly and Touchton 2020). Stadium boosters (e.g. policymakers, team owners, members of local growth coalitions, sports fans) may also contend the venue serves as a public good (Schwester 2007). Public goods are nonrivalrous and nonexcludable,² and when placed within a stadium context, they are typically expressed as civic pride, identity, and fandom (Jiang and Bairner 2020; Johnson and Whitehead 2000; Kellison and Cianfrone 2022; Matheson 2019a). Public spending on infrastructure that provides a public good may be more justifiable because it is ‘likely to be under-produced without a subsidy’ (Johnson et al. 2012, 202).

Past efforts to measure non-pecuniary benefits suggest that despite their lack of attendance at stadium events, non-users may nevertheless derive advantages from a stadium or team. Recently, Matheson (2019a) examined past work on the non-use value of major sports stadiums and found estimates ranged between $17.2–$440 million. To boosters, the fact even non-users stand to benefit from a new or renovated stadium project warrants at least some public spending. To attract broader public support for a stadium project, boosters may also engage in rhetorical strategies that emphasize its multipurpose function (Huberty, Kellison, and Mondello 2016).

Previous strategies to highlight a stadium’s multipurpose function tended to focus on the venue’s capability to support multiple teams (e.g. basketball–hockey, football–baseball), which would do little to change the homogeneity of its users. Somewhere between the user and non-user is a special classification of individuals that utilize services offered at a stadium in a non-consumptive manner (at least in the traditional sense of sport consumer behaviour). For example, when a stadium is temporarily repurposed as a field hospital, emergency staging area, or shelter for those experiencing homelessness or displaced by war, its doors are opened to a much larger portion of the local community (and in some cases, a non-local population). These cases demonstrate a stadium’s potential service to the public at large, as discussed further in the next section.

Adapting stadiums for mass public use

The use of sports stadiums as ERCs is consistent with the guidance of numerous government agencies and advisory groups, including the Federal Emergency Management Agency (FEMA), which has identified stadiums and amphitheatres as prospective high-capacity shelters (FEMA 2006), among other possible uses. The Florida Emergency Preparedness Association defined a mega-shelter as ‘an arena, stadium, convention centre or similar very high-occupancy facility that is used to shelter an exceptionally large population of shelterees from a major disaster’ (as quoted in Florida Division of Emergency Management 2020, D-4). Mega-shelters are further characterized by their capacity to accommodate more than 5,000 individuals, their use during any stage of an emergency (i.e. evacuation, response, recovery), and their location inside or outside of the disaster impact area.

Perhaps the most comprehensive guidance on large, non-traditional shelters is found in the Mega-shelter Planning Guide, published in 2010 by the International Association of Venue Managers (IAVM) in partnership with the American Red Cross. The guide identified several ways in which public assembly facilities may be used as so-called ‘mega-shelters’ in response to large-scale disasters, including as congregate shelters and during shelter-in-place orders. It also described several ways in which a stadium may be used in a non-shelter capacity, including as an evacuation embarkation site, reception processing site, respite centre, disaster recovery centre, staging area, or point of distribution. According to the IAVM, a mega-shelter is not synonymous with a ‘refuge of last resort’ (e.g. Superdome during Hurricane Katrina), which comes with added complexity and liability for the local government.

Several reasons could trigger a stadium mega-shelter declaration (IAVM 2010). First, because stadiums are ‘generally easily identified, recognized, and accessible,’ they may be ‘magnets for those seeking shelter and services’ (10), especially if public transportation and communications infrastructure are severely damaged. Second, and similarly, the effects of public utility shutoffs may be mitigated by a stadium’s backup systems. Third, emergency personnel and services may be consolidated at a single site rather than spread across multiple shelter facilities, which is especially valuable when a large displaced local population is expected to require long-term sheltering.

Much of the information shared in these reports demonstrate the necessity of advanced planning and cooperation between venue operators and local authorities. As noted in the IAVM guide, the oversight of ERCs (in the form of mega-shelters or otherwise) should come from local or state governments because the shelters require extensive coordination between various public and private entities. Furthermore, shelters of this size and magnitude function as small cities and demand the elements of governmental and community support that keep such ‘cities’ safe and well” (IAVM 2010, 7). A review of past stadium emergency response venues further illustrates the need for robust and coordinated planning that pre-empts an emergency.

Hurricane Katrina

The catastrophic effects of Hurricane Katrina on the Gulf Coast are well documented, as is the effectiveness of the mega-shelters activated during and after the storm. In the days and hours before Katrina made landfall, up to 25,000 New Orleanians converged on the Louisiana Superdome, which was infamously dubbed a ‘shelter of last resort’ by the local news media (Greening 2011). Conditions inside quickly deteriorated, especially after the Superdome lost power as the storm raged on. Other problems, like an inadequate supply of basic necessities, were symptoms of the city’s lack of advanced planning. Several scholars, including Kiefer and Montjoy (2006), Eisenman et al. (2007), Grano and Zagacki (2011), and Pfau (2017) have noted the lack of emergency preparedness in the City of New Orleans that led to the poor conditions in the Superdome, and the systemic racial inequities that left Black New Orleanians more at-risk in the face of the storm compared to their White counterparts. Consequently, it has been suggested and well argued that racism played a role in the poor quality of care and support delivered to the evacuees who sheltered in the Superdome, the majority of whom were Black.

The response in Houston, where more than 25,000 residents were evacuated, was decidedly better organized, despite officials only having 12 h to prepare for the first arrivals (Gavagan et al. 2006). The centre of operations was the Astrodome complex (now NRG Park), where evacuees were provided housing, food, and medical care. Several factors contributed to the contrast in care between the Superdome and Astrodome, including the fact that New Orleans was in the direct path of the storm, complicating any last-minute efforts to mobilize an effective emergency response. As Spong (2005) detailed, preparation of the Astrodome mega-shelter was similarly hurried, but local officials were mostly able to meet the challenge:

The best measure you’ll get of how poor the response to Katrina was in New Orleans—no matter which level of government you choose to blame—is how quickly the Astrodome was readied. At three o’clock Wednesday morning, August 31, with New Orleans still filling with water, Harris County judge Robert Eckels was awakened by a phone call from the coordinator of Governor Rick Perry’s division of emergency management (The Dome is under Harris County jurisdiction, making Eckels, the county’s chief executive, the man to call.) By six o’clock Eckels was at the home of Reliant Park’s general manager, and the two of them were working with Mayor Bill White and a host of others to make the relief effort happen. The Astrodome had seen only occasional use in the past six years, and the first step was to get it up and going. That morning the air-conditioning and plumbing were upgraded. While that was happening, the Red Cross shipped in tens of thousands of cots, blankets, and ‘comfort kits’, little bags containing toiletries. …When the first buses arrived at ten that night, the Dome was ready. (261)

Reviews of the Astrodome’s performance reinforce some of the earlier points about a stadium mega-shelter’s potential effectiveness. As Gavagan et al. (2006) found, the sheer size of an empty stadium complex (i.e. more than 100,000 square feet of usable space) made it an ideal space to accommodate a large number of people. Additionally, Houston was not directly impacted by the storm, and thus ‘the regional infrastructure was intact, including transportation, water, electricity, communication, and healthcare’ (937). Stadiums are also built to handle a large assembly of people within a short time, making the Astrodome well-equipped to process both evacuees from New Orleans and the large number of supplies needed for their care (Hamilton et al. 2009). Finally, because the Astrodome was publicly owned, it may have been easier to mobilize resources and accelerate a rapid response. More recently, during Hurricane Harvey in 2017, Houston’s Toyota Center opened as a shelter to alleviate crowding at the nearby convention centre. Acknowledging frustrations with overcrowding at the convention centre, FEMA Administrator Brock Long referenced back to Katrina, asserting, ‘We’re very aware of the issues at the convention center, but let me be clear: This is not the Superdome’ (Associated Press 2017, para. 13).

California wildfires

The 2007 California wildfires required the largest evacuation in the state’s history, including 200,000 residents in San Diego County alone (American Red Cross 2011). Among the more than 50 congregate shelters opened for displaced residents was a mega-shelter at San Diego Stadium (then Qualcomm Stadium and home of the San Diego Chargers). An estimated 10,000 evacuees were sheltered in the stadium and parking lot, although the city prepared ‘to receive up to 100,000 evacuees’ (City of San Diego 2007, 7).

While generally positive, an evaluation of the mega-shelter’s performance after the event found several problems, including a lack of volunteer management protocols, limited capacity to process the massive amount of donated goods, difficulty registering and credentialing evacuees and volunteers, and concerns that undocumented evacuees would be reported to the U.S. Border Patrol (American Red Cross 2011). Despite these issues, city officials concluded ‘the positive association of Qualcomm with successful evacuations and sheltering during the 2007 wildfires’ made it ‘a natural meeting area for those displaced by the next firestorm, if mega care and shelter facilities once again become necessary’ (City of San Diego 2007, 27).

The cost of repurposing stadiums

The Louisiana Superdome sustained significant damage during Hurricane Katrina and underwent a multiphase, $336-million renovation; FEMA awarded more than $172 million for stadium-related projects (Hadley 2009). In Houston, the Astrodome Health Clinic (operated by the Harris County Hospital District) provided care for more than 11,000 patient visits over two weeks at the cost of nearly $900,000 (State of Texas 2007). Following Katrina, the state of Louisiana approved the construction of three standalone mega-shelters for $33 million. These state-owned shelters would reduce the state’s reliance on renting shelter space, which was costing an estimated $3,700 daily (Buchanan 2013). Stadiums used as field hospitals, emergency resource distribution sites, and other emergency functions would also carry costs, though these might not quite be as high as those associated with sheltering people for prolonged periods. In Table 2, we provide additional cost estimates associated with repurposing sports venues for public use.

Table 2. Cost of sports venue operation in public serving functions.

Venue	Location	Owner	Function	Date of Usage	Reported Cost
Houston Astrodome	Houston, TX	Harris County	Temporary shelter for Hurricane Katrina evacuees	September 2005	1,600,000^a
Sullivan Arena and Ben Boeke Ice Arena	Anchorage, AK	Municipality of Anchorage	Homeless shelters	March 2020–present^b	795,180–1,081,351/month^c
Sleep Train Arena	Sacramento, CA	Vivek Ranadivé (private)	COVID-19 alternate care facility	April 2020–present	500,000/month^d
NRG Park	Houston, TX	Harris County	COVID-19 medical shelter	April 2020 (not used)	17,000,000^e
Spokane Veterans Memorial Arena	Spokane, WA	Spokane Public Facilities District	Homeless shelter	May–August 2020	297,000 for 90 days^f
Capital One Arena	Washington, DC	Monumental Sports and Entertainment (private)	Early and election day voting centre	October–November 2020	100,000/day^g
Golden 1 Center	Sacramento, CA	City of Sacramento	California Assembly Organizational Session	December 2020	10,000/day^h

Unless otherwise noted, venue owners are public entities. Reported costs are presented in USD.

^a Cost shared with George R. Brown Convention Center (owned by City of Houston); Office of Congressman Al Green (2006).

^b Ben Boeke Ice Arena closed as an emergency homeless shelter in June 2020.

^c In kind since May 2020; Sabalow and Kasler (2020); van der Meer (2020).

^d Cost varies by number of people served; Dobbyn (2020).

^e McGuinness (2020).

^f Shanks (2020).

^g In kind; Adams (2020).

^h In kind; Nixon (2020).

Cities weighing a public investment in a proposed stadium project must consider how it would serve the public, especially when the funding source comes in the form of general sales taxes rather than more narrow tax pledges or stadium-relate revenues. As part of this calculus, policymakers may consider the venue’s accessibility to an ordinary resident—and, if expected to be low, either reduce the subsidy or ensure greater nonexcludability. With respect to the latter option, maintaining the authority to designate a stadium as a mega-shelter during civil emergencies could reinforce the stadium’s status as a public good. Additionally, not only would this assurance offset costs associated with constructing shelters elsewhere, but it would also compel stadium operators and local officials to establish staff training and emergency planning protocols, thereby enhancing the community’s overall disaster preparedness.

The strategy to incorporate an emergency response function in a stadium’s design when lobbying for public funding is not without precedent. In 1958, Harris County voters approved a $20-million bond to build what would eventually become the Houston Astrodome. Shortly thereafter, county officials began seeking additional funding streams as construction costs grew. Mindful of a potential funding shortfall, they ‘drew up plans for the stadium to serve as a fallout shelter in the event of nuclear war, and submitted this plan to Washington’ (Talbot and Plumer 2015, para. 8). Although the $750,000 cost to add the shelter was ultimately denied, ‘the news stories about the fallout shelter proposal helped promote the idea of the Domed Stadium as a public facility’ (para. 8). Elsewhere, in 1988, Florida legislators attached a provision to a law requiring publicly funded stadiums to provide shelter for unhoused residents on non-event nights, a condition largely ignored until 2012, when Florida senators Mike Bennett and Frank Artiles filed bills to collect refunds from teams that had violated the original homeless shelter provision (Olorunnipa 2012).

We explore a similar strategy below. The unremitting risk of climate change is driving demand for emergency response centers to protect the public before, during, and following extreme weather events (Flavelle 2020). As demonstrated in numerous ways in 2020, sport stadiums can serve a significant portion of the local population, but doing so would require advanced planning and cooperation across public and private entities. Some of this preparation relates to the logistics of supporting persons displaced in the context of disasters and climate change (United Nations 2021). However, a significant amount of planning must also occur during the stadium conceptualization and financing stages. Moreover, ERCs may take many forms, and not all stadiums are suited for use as mega-shelters (due to possible security issues and lack of adequate sleeping arrangements). Instead, the emergency response function may be as a field hospital, resource distribution site, or command centre. In the next section, we outline several important factors to assess when estimating a stadium ERC’s potential value to the public.

Considerations when estimating the public value of a stadium-as-shelter

Demand

Several factors should be considered when determining the public value of a sports facility as an ERC. First, governments would need to examine whether there is actual demand for additional large-scale facilities in the region. Notably, where economic studies of sport have often noted the importance of determining the demand for sport from residents (Borland and Macdonald 2003), governments would need to analyse whether there is a local need for stadiums and to compare the size and capacity of the venue with other existing infrastructure (e.g. schools, airports) that could serve a similar role. If enough infrastructure already exists within the local area to provide proper shelter to residents, it would not be logical to provide funding for a stadium to serve the same function that existing facilities can provide.

Furthermore, any calculations of the value of a sports facility as an ERC should also consider the location and accessibility of the venues. Although there has been a growing trend of moving sports stadiums away from the suburbs and into the downtown areas of cities (Coates and Humphreys 2003; Kellison 2021), an emergency facility would need to be in a critical location that could be easily accessed by residents that need shelter, food or water, medical attention, or other forms of emergency support. In this sense, if a stadium is placed in the middle of a commercial district and is difficult for residents to travel to, its value as a shelter would diminish (Propheter 2020). On the other hand, a venue located closer to residential areas that lacked existing emergency infrastructure could be of greater value in the protection and insurance it provides residents.

The feasibility of a stadium might also be affected by the nature of the emergency to which it is responding. For example, following the COVID-19 outbreak, officials in Wuhan converted local stadiums and other public assembly facilities into temporary (fangcang) shelter hospitals to treat and isolate patients with mild symptoms (Fang et al. 2020). In this case, venues were easily accessible, had configurable indoor and outdoor space, and were centrally located, while still maintaining sufficient distance from ‘high-density residential areas, schools and other densely populated areas in order to reduce the risk of virus transmission’ (2). Thus, even if a stadium can shelter residents, it may not always be appropriate to be utilized in this manner. We discuss this issue further below.

Structural, geographical, and organizational capacity

In light of the potential value provided by sports facilities as an ERC, there is also a need to consider whether a proposed facility is equipped to serve in this capacity. That is, it needs to be determined if the venue would be able to provide sufficient protection and services during a time of emergency, as well as the type of emergencies in which it will be useful to local governments, residents, and other related stakeholders. To strengthen a stadium’s functionality as a mega-shelter, for example, its design should also include space for various needs, including registration; food preparation and feeding; private showers and restrooms; family, children’s, and quiet areas; support staff; service animals; general storage; and health and mental health services (IAVM 2010). In all cases, the privacy and protection of residents should be prioritized, as past cases have shown that these considerations are often sidelined with haphazardly designed shelters. For instance, in the shelter operated at Coliseo Roberto Clemente in San Juan, Puerto Rico, during and after Hurricane Maria in 2017, the limitations of changeroom showers became evident. While changeroom showers (i.e. an open room with several showerheads along the walls at approximately 5 ft intervals) might be appropriate for teams of athletes who all know each other, these fall short of offering acceptable showering facilities for mixed groups of men, women, and children, who may or may not know each other, and who are experiencing extreme distress. For use as a field hospital, a stadium might require an additional set of adaptations, including the provision of critical care receptacles at each bed to ensure reliable use of life-saving equipment.

One essential feature of an ERC is its roof; if the venue is intentionally designed with public safety in mind, an open-air stadium that lacks the capability of being covered during an emergency may render it obsolete for use as a shelter during an emergency. The inclusion of a roof—fixed or retractable—in a stadium’s design is not a minor decision. As shown in Table 3, the roof premium can account for more than one-third of a stadium’s total cost. However, open-air stadiums might be considered for other ERC uses, such as staging areas for emergency vehicles or helicopter landing sites. Considering most modern professional sports fields have sophisticated drainage systems, they should be able to ensure a dry and stable landing area more quickly than other similarly sized plots of land. Such justification was made for using a minor league baseball stadium as a helicopter landing site and FEMA response centre during Hurricane Katrina (Hill 2015).

Table 3. Stadium roof historical costs.

Stadium	Location	Owner	Year Opened	Roof Type	Reported Roof Cost	Reported Total Cost	% of Total Cost
American Family Field	Milwaukee, WI	American Family Field Stadium District & Milwaukee Brewers	2001	Retractable	133	400	33
Highmark BlueCross BlueShield Stadium	Orchard Park, NY	Erie County	—	Fixed	298	1,652	18
Chase Field	Phoenix, AZ	Maricopa County Stadium District	1998	Retractable	70	354	20
Minute Maid Park	Houston, TX	Harris County-Houston Sports Authority	2000	Retractable	39.5	277	14
T-Mobile Park	Seattle, WA	Washington State Major League Baseball Stadium Public Facilities District	1999	Retractable	67	517.6	13
AT&T Stadium	Arlington, TX	City of Arlington	2009	Retractable	250	650	38
Lucas Oil Stadium	Indianapolis, IN	Indiana Stadium and Convention Building Authority	2008	Retractable	70–90	720	10–13
NRG Stadium	Houston, TX	Harris Country Sports and Convention Corporation	2002	Retractable	48	417	12
State Farm Stadium	Glendale, AZ	Arizona Sports and Tourism Authority	2006	Retractable	75	375	21

All venues are owned by a public entity; the ownership of American Family Field is split between the American Family Field Stadium District (public) and the Milwaukee Brewers (private). Reported costs are presented in US$ millions and may include both structural and mechanization systems and, in the case of the proposed new Buffalo Bills’ stadium, an allowance for soft costs.

Sources: AECOM (2021); Frazer (2004); Schoettle (2011).

As discussed in the previous section, site location is also an important consideration when determining the feasibility of an ERC. For example, a stadium site’s location on a floodplain, its proximity to storm surge zones or nuclear power plants, and its susceptibility to soil liquefication following an earthquake may all be relevant factors in the decision to declare (or not) it an ERC (Los Angeles Operational Area 2010). The central location of many stadiums near urban centers may also present problems because they may themselves be located in areas inappropriate for shelters or field hospitals, including places that are within hurricane evacuation zones or flooding isolation (Florida Division of Emergency Management 2020). There is also the consideration of proximity to the people most in need of emergency services, including low-income communities and those with unreliable housing. Consequently, a suburban site or one that is not easily accessible may not be appropriate for all ERC uses.

Indeed, throughout history, there are numerous examples of sports stadiums serving to shelter individuals who have been displaced from their homes. As discussed previously, during Hurricane Katrina, the New Orleans Superdome was used as an emergency shelter. However, because of the lack of services and proper planning, the residents who evacuated to the facility were effectively trapped in the stadium (Gold 2005). The Superdome is estimated to have housed an estimated 25,000 people, but even when sports facilities have hosted smaller groups of people with adequate planning, there have still been issues in having people live in venues that were not properly designed for human habitation. A prime example of this challenge can be found in many of the venues from the 2004 Olympic and Paralympic Games in Greece; they had remained unused for over a decade until they were converted into temporary settlements by the Greek government in response to the influx of refugees fleeing the Syrian civil war. Although the government was able to use these facilities as well as other infrastructure, such as the abandoned Hellinikon airport terminals, to house refugees, they were likewise criticized for not providing adequate conditions for those forced to live within them (Smith 2018). As such, the first thing that needs to be considered in determining the value is whether the venue itself can serve the purpose of an ERC in terms of its design, infrastructure, and services. If a facility cannot do any of these adequately, then it would clearly be the case that subsidizing the venue would not be a good use of funds, which could otherwise be directed to other projects that would be beneficial in times of emergency.

Flexibility of operational control

Beyond the ability to serve as an ERC, there is also a need to consider the scenarios in which a sports venue could be utilized by local governments and ultimately who would have control during emergency situations. As emergency response happens first at the city scale, with support from national and state governments generally following a request from the mayor or a similarly high-positioned local official, the use of city-owned or at least locally-managed sites (in the case of private sports stadiums) is important for the sake of rapid deployment. Compared to other public assembly facilities available in cities like exhibition halls and theatres, stadiums have higher idle rates that generally make them ‘an ideal place for disaster prevention and refuge’ (Song, Ma, and Li 2018, 2). During the COVID-19 pandemic, many sports facilities across the US were utilized as both mass testing and vaccination centers to provide vital services to as large a population as possible (Adler 2021). In certain cases, however, access to these services was temporarily closed because the stadium was hosting another event, such as when Hard Rock Stadium in Miami closed its vaccination and testing centers early because they were hosting the College Football National Championship Game (Burke 2021). Similarly, Atlanta’s Mercedes-Benz Stadium, a federally designated mega vaccination site capable of administering up to 10,000 doses per day, closed on several occasions to accommodate regularly scheduled boxing and soccer matches (Brasch 2021).

Although it would be unlikely that teams would still try to host events during large natural disasters requiring the mass evacuation or sheltering of residents, it would still be necessary for local governments to take control of facilities to be able to prepare and use them as ERCs. If such access and control are not provided, it would again be the case where there may not be value in providing subsidies to these facilities. Of course, the extent to which a government had emergency jurisdiction over a stadium’s operations would be subject to the venue’s ownership and lease agreements. As Miao, Shi, and Davlasheridze (2021) observed, ‘Although local governments are on the frontline of dealing with natural disasters, they are not necessarily best positioned to provide disaster protection even with more fiscal autonomy’ (46). Indeed, many temporary shelters in the US have relied on FEMA’s Public Assistance Program, which may reimburse a facility operator for taking emergency protective measures during a federally declared emergency or major disaster (Federal Emergency Management Agency 2020). Importantly, this program does not aid with lost revenue and fees from cancelled events (Los Angeles Operational Area 2010).

Overall, the key to determining a stadium’s value as an ERC is in its utility to the local community. As we have discussed, this would likely be a function of the demand and need for additional emergency facilities, the location and size of a facility, as well as how easily the public could access it. However, there will likely also be other regional-specific factors that local governments would need to consider regarding whether they should provide subsidies to stadiums to serve as emergency shelters. For example, returning to the case of Hurricane Katrina, because of rising floodwaters, the Superdome was effectively inaccessible other than by boat and helicopter during the hurricane. Although it provided shelter to residents, it also meant that because of its location in an area that was easily flooded, it would be difficult to move evacuees in or out of the building. In this sense, the value a sports venue provides will need to be assessed based on the environment, conditions, and potential hazards unique to each specific region.

It should also be noted that many non-emergent functions may be adopted by stadiums to extend their non-excludability beyond emergency situations. As mentioned previously, in 2020 and 2021, several professional arenas and stadiums in the US were used as polling stations in the federal election (Orr and Kellison 2020) and as vaccination sites in the early stages of the COVID-19 vaccine rollout (Sears, Cianfrone, and Kellison 2021). The use of stadiums for these other public functions can further instil trust in the facility among community members and may further aid in strengthening the working relationships between stadium officials, emergency service providers such as FEMA and the National Guard, and local authorities (Fischl et al. 2022), making ERCs easier to activate if and when necessary.

Conclusion

We can make two predictions about American cities with confidence: they will continue to (1) build stadiums and (2) grapple with the effects of climate change. With respect to the latter prediction, cities will not only need to adapt their infrastructure to mitigate damage from extreme weather events but also respond to large-scale emergencies when they occur. As Donahue (2014) wrote, financially, this response is a public responsibility:

Governments are charged with helping citizens protect themselves from the threat of disaster, responding to protect lives and property when disaster strikes, and providing assistance that promotes the return to normalcy for individuals, businesses, and communities in the wake of a disaster. The cost of meeting these needs falls to taxpayers. (100–101)

If cities are to continue building stadiums and funding climate adaptation and emergency response efforts, it would be logical to find overlap in these pursuits.

In nearly all cases, the cost of building stadiums also falls to taxpayers—at least in part. These stadium subsidies are usually met with some public diffidence, particularly by those who challenge the contention that a professional sports stadium is beneficial to the public at large. In this article, we have situated this argument in the context of climate change and its impact on urban spaces and human settlements. As discussed above, at least some portion of a stadium subsidy may be justified if city leaders assured the public that the stadium would double as an ERC during emergencies. For example, the cost to subsidize a retractable roof (and operate it) might be more tolerable if a city’s emergency needs were met.

The added cost of designing a stadium to meet mega-shelter or field hospital standards would be relatively low (with the possible exception of a roof structure). Modern engineering advances mean most stadiums already have the structural capability to safeguard residents during a public emergency. Instead, changes would be required to the way publicly owned stadiums are operated and managed. The conditions that would trigger an emergency takeover; the legal liability of patients or residents and ensuring their safety; and the cost of operating the ERC (and event cancelations; Murfree and Moorman 2021) would all require codification. Previous attempts to define such provisions (e.g. requiring sustainable design certification, community benefits agreements) have failed to gain widespread traction. While policymakers may benefit from such an agreement, there may be little incentive for stadium operators—often team owners with the leverage when negotiating with local government—to concede control of the stadium. Still, global cities are increasingly investing in efforts to reduce their climate vulnerability. As the public cost of these mitigation and adaptation strategies is similarly expected to grow, a proactive stadium ERC declaration would not only complement a city’s broader climate (and economic) policy but also strengthen the stadium’s status as a public good.

Acknowledgements

The authors thank Geoffrey Propheter, Daniel Rascher, and Tiffany Richardson for their constructive reviews of previous versions of this article.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 In this paper, we use stadium as a catchall term to include all major sports venues, including, but not limited to, indoor and outdoor arenas, ballparks, and speedways.

2 Our application of the term nonexcludable to stadiums being used as public shelters is intentional—an effort to evoke the concept of the stadium as a public good. Of course, a stadium mega-shelter is nonexcludable insofar as it can meet human demand (in size and resources), and we acknowledge there are clear circumstances in which a stadium’s capacity would be exceeded by human demand, especially following a disaster. A more academic debate on this issue may centre on congestible public goods (Ahn, Isaac, and Salmon 2009), club goods (Cornes and Sandler 1996), and expanded typologies of goods (Ostrom 2010).