A long tradition of research on racist attitudes and discriminatory practices has shown that threatening events – such as terrorist attacks, acute economic shocks and outbreaks of infectious diseases – are associated with greater hostility among racial and ethnic majority group members (‘in-groups’) towards immigrants or ethnic minorities (‘out-groups’).
Coverage of threatening events in social and traditional media plays an important role in this link. The highly threatening nature of Covid-19 itself, as well as the lasting negative economic, political and social consequences of the current response to the pandemic, are likely to increase xenophobic and racist attitudes and associated hate crimes and incidents in the short term, with potentially long-term consequences too.
It is important to keep in mind that while we discuss general patterns in racist and anti-immigrant attitudes and behaviour of the dominant group towards ethnic, racial, religious and national minorities, there is variation over time and across groups (Ford 2008; Poynting and Mason, 2007).
As Figure 1 shows, hate crime reports have been generally increasing over the past seven years. Some threatening events, such as the referendum result in June 2016 and the terrorist attacks of 2017, match with sharp spikes in hate crime reports; but other events, such as the Charlie Hebdo shooting in Paris in 2015, do not.
It is also difficult to separate the effect of threatening events from unrelated trends or events that may influence hate crime reporting. Recent social science research that measures this relationship in more robust ways is summarised below.
Figure 1: Number of racially or religiously aggravated offences recorded by the police by month
Source: Home Office, 2019
What does research in economics and sociology tell us about the association between threatening events and racist attitudes and behaviour?
Theories of social or group-based identities, developed by social psychologists, have long established the existence of favourable attitudes and behaviour towards in-group members and discriminatory practices towards out-group members (Tajfel, 1981).
Sociologists (such as Bobo, 1999) and economists (Akerlof and Kranton, 2000) discuss how these individual preferences, produced and strengthened by group-based identities, operate in the aggregate to shape ethnic inequality in life chances, as well as perpetration of and exposure to discriminatory acts. In times of increased competition for scarce resources, as well as opportunities and events that heighten the salience of social identities, these prejudicial attitudes and behaviour may strengthen.
Fear of infectious disease is also associated with irrational economic and social responses (Smith, 2006; O’Shea et al, 2020, on US data). Research finds that infectious disease outbreaks, such as the Ebola crisis in 2014, led to an increase in racist and xenophobic attitudes (Broom and Broom, 2017, on Australian data; Kim et al, 2016, on US data). The link between increasing saliency of infectious disease and anti-immigrant attitudes appears to hold even in experimental settings (Bartos et al, 2020, on data from the Czech Republic).
Economic downturns may also be associated with an increase in discrimination against ethnic and religious minorities: research finds substantial increases in both self-reported racial prejudice, and wage and employment gaps between majority and minority members during periods of higher unemployment (Khattab and Johnston, 2013, on UK data; Kingston et al, 2015, on data from Ireland; Johnston and Lordan, 2016, on UK data).
Information channels play an important role. Racist and anti-immigrant media portrayals of threatening events are associated with increases in hate crime and xenophobic and racist behaviour (Adena et al, 2015, on data from Germany). This association can be intensified where such portrayals are spread via social media (for a review, see Zhuravskaya et al, 2020).
Research suggests that threatening events involving one out-group can have wider impact, increasing xenophobia and racist attitudes and behaviours more generally. For example, during the Ebola outbreak, which originated in West Africa, researchers found increases in xenophobic attitudes both directly towards West Africans, but also towards undocumented immigrants, the majority of whom were not of African descent (Kim et al, 2016, on US data).
Such ‘spillover’ effects have also been shown to relate to media portrayals. For example, researchers have demonstrated that negative media portrayals linking Islamophobia and immigration following the 9/11 attacks (Romero and Zarrugh, 2018, on US data) resulted in longer prison sentences for minority group members perceived as immigrants, Hispanic defendants, although not for minority group members perceived as native-born, black defendants (McConnell and Rasul, 2020, on US data).
Spikes in hostile attitudes and hate crimes against ethnic minorities and immigrants are commonly observed following terrorist attacks, highly publicised inter-racial crimes and contentious political events such as Brexit (Home Office, 2019, on the UK; Jäckle and König, 2018, on Germany; King and Sutton, 2013, on the United States; Legewie, 2013, on European data; Disha et al, 2011, on the United States; Poynting and Mason, 2007, on the UK and Australia).
It is less known how long such increases in hostility endure. For example, in the case of Brexit, anti-immigrant attitudes have already declined to below pre-Brexit levels (Schwartz et al, 2020).
What new work is emerging on changes in xenophobic and racist behaviour and attitudes due to the Covid-19 crisis specifically?
The emerging evidence shows that the Covid-19 pandemic is associated with an increase in hate crime reporting, especially against Chinese and East Asian minorities. Research also shows that the crisis may increase hostility towards out-group members more generally. Social media use, especially exposure to far right-wing propaganda, may exacerbate this effect.
In the UK, the United States and Canada, police data and reports from non-governmental reporting organisations reveal dramatic increases in reporting of hate crime against East Asian minorities (Tessler et al, 2020; BBC News, 2020; Vancouver Sun, 2020). Police forces across the UK received 267 incidents of hate crime reported by Chinese minorities in the UK during the first quarter of 2020, compared with 375 incidents during the entire year in 2019 (Sky News, 2020).
New research using a money allocation experiment of a nationally representative sample in the Czech Republic under lockdown finds that increasing the saliency of Covid-19 increased hostility towards immigrants, such that the experimental subjects allocated less money to non-Czech participants when Covid-19 was made salient (Bartos et al, 2020).
There is emerging evidence that increases in out-group hostility are linked to media portrayals of Covid-19. Web analysis finds increases in anti-Chinese language both on fringe and mainstream (Twitter) websites (Schild et al, 2020). Survey research from the United States establishes a link between social media use during the Covid-19 pandemic and xenophobic attitudes specifically towards Chinese minorities (Croucher et al, 2020).
Who is likely to experience an increase in the risk of racist attacks and harassment due to Covid-19?
There is emerging evidence (see also above) that East Asian minorities are most at risk of racist attacks. For example, disproportionately high declines in mental health during the Covid-19 crisis experienced by Canadian residents of East Asian origin might be partially explained by experiences of discrimination (Wu et al, 2020).
Research also suggests spillovers of Covid-19-related hostility to immigrants more generally, as evidenced in experimental work from the Czech Republic (Bartos et al, 2020) as well as the anti-immigrant (as well as anti-Chinese) rhetoric employed by the US president and recent immigration restrictions introduced in the United States.
There is evidence showing that minorities living in areas with a higher proportion of their own ethnic group have a lower risk of experiencing ethnic and racial harassment (Nandi and Luthra, 2016; Disha et al, 2011, Becares et al, 2009; Dustman et al, 2004).
Evidence is mixed on whether this is because minorities are less likely to come into contact with majority group members, because majority group members living in such areas are less hostile or because the potential sanctions against ethnic and racial harassment is higher in areas of higher minority concentration.
Economically advantaged minorities are more likely to report ethnic and racial harassment, which could be due to greater confidence and ability to identify incidents as being due to racism and discrimination, or due to greater contact with majority group members (Dixon et al, 2010; Steinmann, 2019; Verkuyten, 2016).
UK evidence shows that women are less likely to report experiencing ethnic and racial harassment, but more likely to find public places unsafe and avoid them (Nandi and Luthra, 2016).
How reliable is the evidence?
The reliability of this data-based evidence depends on:
- The accuracy of reporting of attitudes by the dominant or majority group towards ethnic, racial, religious and national minorities.
- The accuracy of reporting of experiences of discrimination and harassment by ethnic, racial, religious and national minorities.
- The robustness of the statistical methods used and the representativeness of the sample.
A vast body of research in survey methodology has established the role of social desirability bias in survey reporting, where survey respondents have a tendency to report in accordance with social norms, which is heightened in the presence of an interviewer (Lavrakas, 2008; deLeeuw, 2005; Tourangeau et al, 2000). As a result, there may be under-reporting of racist and anti-immigrant attitudes in societies where such attitudes are considered to be unacceptable.
In surveys, self-reporting on experiences of discrimination and harassment is hindered by the inability to identify such acts (for example, because the person is new to the region and not familiar with racial slurs, etc. or because such incidents are so common that they have become normalised) and lack of confidence in reporting.
In the case of reporting to organisations and institutions including the police, not only are these barriers more difficult but there is an additional hurdle: lack of information on how and where to report. It should also be noted that by definition, only those incidents that qualify as crimes are counted as hate crimes: the others are recorded as hate incidents.
The reliability of survey data-based evidence depends on how well the population is represented in the sample from which the survey data are collected, and the statistical methods used to produce the estimates. Weighted estimates based on surveys of samples drawn from a specific population (such as young people or specific minority groups) will provide estimates that are only generalisable to that population.
Finally, it may also be difficult to establish the existence and strength of the causal link between a threatening event and changes in racist and anti-immigrant attitudes and discriminatory behaviour due to confounders.
For example, a threatening event may coincide with increases in immigration or deterioration in inter-group relations: it is therefore difficult to determine whether any increase in racist attitudes is due to the event or related trends in attitudes. In the case of the current pandemic, this may not be a problem as it is an unanticipated event that could not have been the result of racist and xenophobic attitudes or behaviour.
Still unknown: the effects of Black Lives Matter and the lifting of lockdown
What is different about the current crisis compared with threatening events in the past is the shutdown following the outbreak. So while hate crimes may show a drop during lockdown, they may have spiked in online spaces where social interaction has shifted. The limited evidence that was available before lockdown in the United States and the UK (Tessler et al, 2020 and Sky News 2020) indicates that there is a possibility of rise in hate crimes against residents of Chinese and East Asian origin.
But something else happened during the Covid-19 crisis: the differential impact of the disease on minority communities – due to their concentration in key worker and healthcare occupations, and their higher rates of infection and death – was widely reported in the UK and the United States (PHE report, 2020; Platt and Warwick, 2020; Meer et al, 2020).
This work also highlights the existing economic vulnerability of racial and ethnic minorities (Platt, 2020), and the role of existing racist structures and discriminatory practices (Becares and Nazroo, 2020; Guardian, 2020), which were widely discussed.
This media attention was immediately followed by the unprecedented rise and spread of the Black Lives Matter (BLM) movement across the United States and the world following the death of George Floyd, an unarmed black man in the custody of the Minneapolis police. Protests around this movement are continuing in the UK at the time of writing with responses from both private and public institutions as well as policy-makers.
Given the potentially countervailing events of the Covid-19 crisis, the decrease in in-person interaction during lockdown, and the heightened awareness of discrimination and inequality exemplified by Black Lives Matter, it will be difficult for researchers to unpick the causes of changes in hate crime and discriminatory attitudes arising during this time.
Read the original blog on the Economics Observatory website here.