The spread of Covid-19 and attending football matches in England
The rapid spread of the Covid-19 virus meant that by mid-March 2020 the government had stopped outdoor sports from being played in England. Since then, football has resumed behind closed doors with no fans in the stadiums. Whether football fans should be allowed to return to attending matches is now the subject of much debate. There is research showing that indoor events appear to increase the spread of the virus (Ahammer et al. 2020), but what about outdoor events?
Football in England provides us with a rich setting for data to allow us to research these effects. Every year between August and May, millions of fans attend football matches in their local area and around 10% also travel as ‘away’ fans. These fans are usually sectioned off in a separate part of the ground, but often travel on the same trains or buses and access the same bars and restaurants as home fans. Furthermore, upon arrival at the ground, fans in larger stadiums will often enter via a turnstile system before being filtered in and out of the stadium through concourses where they also often congregate at half-time. If the virus is more likely to spread indoors than outdoors, does it not also seem likely that all these indoor interactions associated with football viewing will lead to an increase in spread of the virus?
What of past research on virus spread, economic impact and in particular relating to sport? There is literature on the economic impact of a public health emergency such as through H1N1 or influenza and also the subsequent impact on industries like tourism (Keogh‐Brown et al. 2010, Kuo et al. 2009, Page et al. 2012, Rassy and Smith 2013, Adda 2016). There is a vast literature on the demand for attendance at professional sports (Buraimo and Simmons 2008).
Combining these two literatures, Gitter (2017) shows that H1N1 reduced attendances at Mexican baseball games by up to 30% and this reduction was even greater with influenza added on top of that. Likewise, Cardazzi et al. (2020) show that in the US, having a sports franchise in a city was consistent with 4-24% chance of flu deaths within that area. In addition, during the early stages of Covid-19, research looked at what happened to attendances in Belarus, where football was allowed to continue (Reade et al. 2020a, Reade et al. 2020b). The authors show that demand clearly remained for viewing sport in the midst of a pandemic, even if at a lower level than before. Although this past research has examined various aspects of the economic demand for football and the effect of virus spread on attendances, our recent paper seeks to address the gap in the literature of whether attending matches helped increase the spread of this airborne virus (Olczak et al. 2020).
We compiled a database of football matches from the 2019/20 season, covering the nine tiers of football leagues (the four top divisions plus the network of non-league competitions). We could then exploit the fact that until the government stopping outdoor sports in mid-March, attendance at football matches across England carried on despite Covid-19 cases and deaths increasing. Matches that took place include the League Cup Final at Wembley and the European Champions League tie between Liverpool and Atletico Madrid. More broadly, matches up and down the country involving both large and much smaller clubs continued to take place with, for example, a total of 180 matches took place on the 7 March and 42 on the 14 March.
Figure 1 Distribution of % of capacity filled in 2019/20 season
We also collected data on Covid-19 cases, deaths and excess death rates by local authority area and a range of demographic indicators. This allowed us to analyse whether different levels of fans attending matches across these local areas had a lagged effect on the case or death rate. Furthermore, given that many fans travel across the country to see their team play away from home, we estimate the effect on cases and deaths in the local area of the away team as well as the home team.
Overall, we found that there were strong effects of attendance at matches on Covid-19 cases and deaths before we added the demographic control variables such as population density, age, and ethnicity. Adding these controls reduced the effect slightly but it was still significant. We found that an extra football match resulted in an additional six Covid-19 cases and three excess deaths per 100,000 people. Furthermore, once the controls were added, the impact on cases and deaths did not vary by stadium capacity utilisation – a stadium being 20% full had the same impact as if it were 90% full.
Figure 2 Covid-19 confirmed case rates by area by number of matches in March
Figure 3 Summary of regression results of football match activity on mortality in the first half of 2020 in England
Notes: Each column of plots relates to a measure of mortality (Covid-19 cases, Covid-19 deaths, and excess deaths), the top row is without any control variables, and the bottom row adds control variables. Solid dots are regression coefficients on the footballing measure, and circles are the upper and lower confidence intervals (90% signicance level).
When should fans be allowed to return to matches?
The economic impact of playing football behind closed doors on teams outside the English Premier League has been devastating. Whilst teams in the Premier League are cushioned by the television revenues they receive, smaller clubs are largely reliant on gate receipts from season ticket and on-the-day turnstile sales plus the commercial revenue from concessions at the stadium. The fear is that teams will be financially unable to continue. Despite this, our research suggests that there should be extreme caution in allowing fans to return to matches. While there is evidence that fans socially distance voluntarily, our findings suggest that the standard way in which fans congregate to attend matches can facilitate the spread of the virus, even if capacity utilisation in the stadium is low. Certainly, fans will need to act in a more cautious manner in future whilst attending matches. Government pilots to let fans back in had tentatively begun in England but have been hampered by national and local restrictions being introduced. If these pilots are to restart, they will need to be closely monitored to see whether spreading fans out across stadiums and timing of entry into the stadium can effectively prevent the spread of the virus spread.
Adda, J (2016), “Economic activity and the spread of viral diseases: Evidence from high frequency data”, The Quarterly Journal of Economics 131: 891-941.
Ahammer, A, M Halla and M LACKNER (2020), “Mass gatherings contributed to early COVID-19 spread: Evidence from US sports”, Working paper.
Buraimo, B and R Simmons (2008), “Do sports fans really value uncertainty of outcome? Evidence from the English Premier League”, International Journal of Sport Finance 3.
Cardazzi, A, B R Humphreys, J E Ruseski, B Soebbing and N Watanabe (2020), “Professional Sporting Events Increase Seasonal Influenza Mortality in US Cities”, Available at SSRN 3628649.
Gitter, S R (2017), “The H1N1 Virus and Mexican Baseball Attendance”, Athens Journal of Sports 4.
Keogh-Brown, M R, S Wren-Lewis, W J Edmunds, P Beutels and R D Smith (2010), “The possible macroeconomic impact on the UK of an influenza pandemic”, Health Economics 19: 1345-1360.
Kuo, H-I, C L Chang, B-W, C-C Chen and M McAleer (2009), “Estimating the Impact of Avian Flu on International Tourism Demand Using Panel Data”, Tourism Economics 15: 501-511.