AdobeStock_296714171.jpeg
VoxEU Column COVID-19 Frontiers of economic research

Social pressure in football matches: An event study of ‘remote matches’ in Japan

Our behaviours are highly influenced by social pressure. This column takes as a natural experiment the 2020 season of the Japanese professional football league, which held matches without spectators due to the COVID-19 pandemic, to examine whether the presence of spectators puts pressure on referees’ decisions. The authors find that the home team advantage is real: the number of fouls decided against the home team decreased significantly in matches with spectators. The absolute number of home-team supporters mattered.

Our behaviours are highly influenced by social pressure. For instance, amid the COVID-19 pandemic in Japan, many people stayed at home and wore face masks, following the government’s ‘requests’ even though they were not legally binding. Such behaviours may be partially caused by social pressure in addition to the primary purpose of infection prevention. 

The relationship between social pressure and people’s behaviour has been one of the main concerns in economics since the pioneering work by Akerlof (1980). One strand of the literature has analysed this relationship by focusing on the presence of spectators and match outcomes in professional sports (e.g. Endrich and Gesche 2020, Bryson et al. 2021, Scoppa 2021). We follow this literature (Morita and Araki 2021) and take as a natural experiment the 2020 season of a Japanese professional football league, which was unexpectedly forced into holding matches without spectators due to the COVID-19 pandemic, to examine whether the presence of spectators puts pressure on referees’ decisions.

‘Remote matches’ in Japan

In Japan, a match with no spectators due to the pandemic is called a ‘remote match’. The 2020 season started on February 21 and only the first period of the season was held as usual. After the first period, the season was suspended until June due to the spread of COVID-19. Matches resumed on July 4 for the first division (J1 league) and June 27 for the second division (J2 league) but were held with no spectators, and the situation continued in the next period in early July. As a result, 43 out of 768 games (5.6% of all games) were played as remote matches. This meant that – based on the 2019 season – an average of 21,000 spectators per match in J1 and 7,000 spectators per match in J2 were removed from the stadium during this period. By taking advantage of this unprecedented situation, we analyse whether the referees’ decisions, such as on fouls and yellow cards, are affected by the social pressure of spectators.

After 10 July 2020, the restriction on the number of spectators was relaxed in stages depending on the infection situation, as shown in Table 1. Broadly speaking, two periods after resuming matches were held completely as remote matches, and then for a while, only home-team supporters were allowed to watch the games in the stadiums under a tight attendance cap (step 1). After a transition period, away-team supporters were also allowed to enter stadiums and the attendance limit was further relaxed (steps 2 and 3). 

In summary, during the first step, the number of spectators was small but the stadium was filled 100% with home-team supporters. In contrast, during the second and third steps, the absolute number of home-team supporters in the stadium increased but their ratio was decreasing due to the mobilisation of away-team supporters. We focus on the differences in the limitation of spectators and identify whether the referees exhibited bias and whether the source of any such bias is the absolute number of home-team supporters or the proportion.

Table 1 Timeline of restrictions on spectators

 

Referee bias

We use difference-in-differences to estimate the impact of home-team supporters on the referees’ decisions. This method emphasises the benefits to the home team – in the form of referee decision-making – arising from the presence of home-team supporters. It captures the difference between the difference in referees’ decisions for the home team and away team during matches (i) without spectators and (ii) with spectators (Figure 1).

Figure 1 Difference-in-differences

 

Notes: This figure graphically shows the difference-in-difference estimation. Difference (i) corresponds to the difference in the number of fouls for home versus away teams in matches without spectators, while difference (ii) corresponds to the difference in the matches with spectators. Our interest is in the difference-in-differences, corresponding to (i)-(ii).

The analysis reveals that the number of fouls decided against the home team in matches with spectators decreased significantly, by about 1.05, indicating that the home-team advantage is statistically significant although its magnitude is small. On the other hand, the number of yellow cards received by the home team was unaffected. Moreover, the importance of the absolute number of home-team supporters was evident based on the stadium capacity restrictions: we observed significant home-team advantages in the referees’ decisions when the matches were played under the relatively lax restriction on the audience cap, with the away-team supporters. However, only about 1.38 fouls were detected as the advantage in the referees’ decisions to the home team.

Conclusion

We examined the impact of social pressure on human behaviours by exploiting the natural experiment of professional football league matches with no spectators under the COVID-19 pandemic. Relying on information about detailed differences in spectator restrictions, we extracted the restrictions’ effects with a reliable identification strategy. Our work is an empirical analysis that supports the classic economic theory of social behaviours, which represents an uninterrupted line of literature from the pioneering work of Akerlof (1980). 

The power of social pressure is not particular to the referees’ decisions in sports matches: it can also apply to peer pressure under the COVID-19 pandemic that forces people to wear face masks and refrain from going out. Therefore, it is desirable to continue to clarify the relationship between social pressure and people’s behaviour through the assessment of such events.

Editor’s note: The main research on which this column is based (Morita and Araki 2021) first appeared as a Discussion Paper of the Research Institute of Economy, Trade and Industry (RIETI) of Japan.

References

Akerlof, G A (1980), “A theory of social custom, of which unemployment may be one consequence”, Quarterly Journal of Economics 94(4): 749.

Bryson, A, P Dolton, J J Reade, D Schereyer and C Singleton (2021), “Causal effects of an absent crowd on performances and refereeing decisions during Covid-19”, Economic Letters 198: 109664.

Endrich, M, and T Gesche (2020), “Home-bias in referee decisions: Evidence from ‘ghost matches’ during the Covid19-pandemic”, Economic Letters 197: 109621.

Morita, H, and S Araki (2021), “Social pressure in football matches: An event study of ‘remote matches’ in Japan”, Applied Economic Letters (forthcoming).

Scoppa, V (2021), “Social pressure in the stadiums: Do agents change behavior without crowd support?”, Journal of Economic Psychology 82: 102344.