Dr Joel Carr is a researcher at the Netherlands Centraal Planbureau (Bureau for Economic Policy Analysis) in the environment programme. His research at the CPB includes microeconometric analysis in topics such as mobility, income, and energy consumption. During his PhD he wrote his dissertation on causes of hate crime, looking at the effects of Brexit, COVID-19, and the 2020 Black Lives Matter protests. The views and opinions expressed in this blog are those of the speakers and do not necessarily reflect the views or positions of CPB.
Dr Joanna Clifton-Sprigg is a Lecturer in Economics at the University of Bath. Her research interests are in labour and gender economics, migration and economics of education. Her recent research addresses policy-relevant questions, among them the causal effects of shock events (e.g. Brexit, COVID-19 and Ukraine war) on hate crimes, social cohesion and migration.
Dr Jonathan James is a Senior Lecturer in the Department of Economics at the University of Bath. Jonathan's research interests are in the field of applied microeconomics, with a focus on applied policy orientated research.
Prof. Sunčica Vujić is a Professor of Applied Econometrics at the University of Antwerp. She holds affiliations with the VU University Amsterdam and the University of Bath. She is a Research Fellow at the IZA – Institute of Labor Economics and the Global Labor Organization (GLO). Her research interests are in the fields of applied econometrics and labour economics, focussing on crime, education, gender, discrimination and pro-social behaviour. She serves the community as a secretary of the Executive Committee of the European Society of Population Economics (ESPE). She holds a Ph.D. in economics from the VU University Amsterdam.
This is a big year in politics, with more people going to the polls than any other. With European Parliament elections and general elections in many countries, including India and South Africa, already behind us and with elections coming up in the UK, France and the USA – we are being asked to make our preferences known. In the run up to the vote we are watching strong-worded politics. In the UK the televised, heated party head-to-heads are a perfect example of that.
For years now immigration has been a running theme. This election is no exception. Already in the first TV debate Rishi Sunak and Keir Starmer were quizzed about the Rwanda bill and the net migration target, which the Reform party pledges to set it to zero.
This should come with a clear warning: What leaders say in public and how they say it matters. A divisive, hateful debate around immigration or ethnic minorities can lead to hate crimes and will be amplified by (social) media.
We should care about this because hate crimes bring social and economic costs to victims and the society. The immediate consequences one thinks of include physical and psychological harm to the victims, their families and the broader community, costs of property damage, hospitalisation, policing and to the judicial system when cases are brought to court. Hate crimes also affect social cohesion through erosion of trust.
Political events are a culmination of views and can influence attitudes (and acts of hate).
Previous research shows that political events cause changes in attitudes and hate crime patterns. For instance, Leonardo Bursztyn, Georgy Egorov and Stefano Fiorin show in their work that Donald Trump’s rise in popularity and his 2016 election victory increased the willingness of people to express anti-immigration views. The social acceptability of these views also increased. The event has also coincided with an increase in hate crimes in the US.
Local elections can have a significant impact on local hate crimes too. Focusing on Italy, Alessio Romarri shows that the likelihood of hate crime being committed is higher in the municipalities in which an extreme-right mayor narrowly won the elections compared to areas with a marginal defeat of a far-right candidate. These examples demonstrate that information shocks from elections can be important factors in hate crime occurrence.
Our work on the Brexit referendum and on the COVID-19 pandemic shows that unexpected, divisive events can cause a spike in hate crimes. Like Romarri, we find that areas which narrowly voted to Leave experienced a significantly larger hate crime shock than areas that marginally voted to Remain. Meanwhile, there was no discernible increase in Scotland which voted at the national level to Remain.
Immigration featured heavily in the referendum campaign and Chris Prosser, Jon Mellon and Jane Green show that it also mattered most to the Leave voters. The referendum outcome itself came as a shock to many and was in stark contrast to polling and betting data. Given the heavy immigration focus of the Leave campaign, the Vote to Leave the EU could be seen as a signal of social acceptance of xenophobic beliefs and behaviours, even if this was not the intent of the campaign and voters.
This new perception of social acceptability of anti-immigration views would in turn lower the perceived costs of committing a hate crime against racial and religious minorities, particularly Eastern Europeans. Therefore, it had the potential to incite hate crime.
In Figure 1 we show the various spikes in racial and religious hate crimes reported to the UK police forces caused by significant, unexpected events occurring in the UK and around the world between January 2013 and December 2016, the period for which we have data. Events which trigger a clear response in terms of hate crimes vary – from terror-related events (murder of Lee Rigby, terrorist attacks) through to elections. The spike following the Brexit referendum is visibly higher, on-par with the reaction following the murder of Lee Rigby.
Notes: Data used are from the Freedom of Information Requests sent to Police Forces in the UK, from Carr et al (2020). The dots in the plot represent the estimates of the size of the effect a given event had on racial and religious hate crime (RRHC) in the UK. The whiskers around them represent the 95% confidence intervals. When the whiskers cross the horizonal axis, we cannot rule out there being no effect. The events which caused a clear spike in RRHC were the murder of Lee Rigby, ISIS gains in 2013, ISIS murder of western journalists and aid workers, Brexit referendum. These estimates were obtained in an event study approach comparing changes in racial and religious hate crimes over time, relative to a chosen baseline period.
In case of the COVID-19 pandemic we show that racial hate crimes against East Asians in the UK increased significantly throughout 2020, beginning with the first UK confirmed cases and travel ban to mainland China. Having analysed the relationship between hate crimes in the UK and tweets linking COVID-19 and certain nationalities, we argue that the explicit link made between the virus and its origins in China directed the hatred towards people with East Asian origins.
(Social) media often fuel divisive debates but could be used for a good cause instead.
Shock events are accompanied by heavy (social) media coverage and open debates. Controversial and divisive topics fly high in (social) media. In particular, with respect to anti-immigrant debates (such as the Rwanda plan or illegal channel crossings in the UK), two points could be made. Firstly, the salience of negative coverage of minority group can spread. Secondly, fake news spread quicker in social media than facts.
Social media make coordination and collective action easier. They can also change attitudes through exposing individuals to views of others. Moreover, they can normalise controversial or unacceptable views because of the existence of echo chambers. Namely, social media can distort the perception of social norms by reducing the users’ exposure to different parts of society online.
Research shows that (social) media coverage can influence behaviour, particularly in the context of polarised and hateful discussions. For instance, when events are portrayed in a racist and anti-immigrant rhetoric, they trigger hate crime and xenophobic and racist behaviour. Mueller and Schwartz show that the post-Trump election hate crime increase was greatest in areas with higher Twitter use.
This suggests that social media can mediate hate crime shocks caused by elections and campaigns. Ria Ivandic, Tom Kirchmaier and Stephen Machin demonstrate that anti-Muslim hate crime surges occurring in the days following terrorist attacks were further amplified by real-time media. Schneider-Strawczynski and Valette show that increased media coverage of immigrants caused a polarisation of the attitudes.
However, just as research finds that negative coverage contributes to an increase in hate crime, positive coverage or information on minority racial or religious groups can reduce hate crime.This is exemplified in the case of the footballer Mo Salah, whose visibility in Liverpool reduced local racial and religious hate crime. Similarly, Zussman shows that the outbreak of COVID-19 in March 2020 led to a decline in discrimination against Arab doctors in Israel and that media coverage of the role these doctors played in the fight against the virus positively contributed to it. At the same time, he demonstrates that ethnic riots in Israel in 2021 increased discrimination against the same group.
In our work we demonstrate that any coverage of the hate crime triggering event can amplify outcomes. Having shown that the Brexit referendum caused a spike in racial and religious hate crimes in the UK, we show that print and social media coverage further contributes to this increase. We focus on print articles related to two topics: Brexit and hate crimes, published over time in the run up to and after the referendum. We show that as Brexit and hate crimes became more salient in society (as measured by media coverage), racial hate crime rates increased, albeit with a delay. Similarly, looking at tweets related to these two topics, we show that social media coverage is also positively correlated with hate crimes and seems to take effect sooner than that of printed media.
In case of the COVID-19 pandemic, we show that tweets about the number of COVID-19 cases in China and the salience of lockdown in the UK are positively correlated with hate crimes against East Asians. On the other hand, #coronaracism tweets signalling objections to hate crimes against minorities correlate negatively with the number of recorded hate crimes. Interestingly, tweets about the number of COVID-19 cases in the UK are not correlated with hate crimes. This, once again, highlights that (social) media can contribute to an increase in hate crime but can also help reduce it.
Post-event actions by public figures can add to or mitigate the spike in hate crimes.
Politicians and public figures can play a positive role in counteracting hate crime, just as the #coronaracism tweets signalling disapproval for hate crimes did during the COVID-19 pandemic. This can be seen in case of the UK in 2016 and the response of leaders in these countries to post-election hate crime shocks. In the UK the mainstream politicians across the board publicly denounced the hate crime spike and by late summer 2016 the Crown Prosecution Service announced that an uplift would be applied to sentencing for hate crimes.
What you say and how you say it in public matters
Our findings bring forth important lessons for policymakers and society. First, careful attention must be paid by the media and politicians to not associate or blame a specific group of people or country for an unexpected, negative event. We focused on Brexit, US elections and COVID-19 pandemic but this can be extrapolated to other current events such as the wars in Ukraine or Gaza. These conflicts could lead to increases in hate crimes against specific ethnic groups in western nations. Second, when signals are sent that foreigners pose a risk to the society, there must be strong message given by politicians, police, and private citizens that xenophobic expressions and racial hate crimes will not be tolerated. Research has made it clear that signals play an important role in the hate crime-decision process. Therefore, to prevent any future hate crime shocks, consideration should be given to how the signals sent and words used are interpreted by potential offenders.
The views and opinions expressed in this blog are those of the speakers and do not necessarily reflect the views or positions of CPB. All articles posted on this blog give the views of the author(s), and not the position of the IPR, nor of the University of Bath.
Respond
