Evidence to the Speaker’s Conference on the security of candidates, MPs and elections

Our election research over recent years has identified a trend of high levels of abuse and intimidation of candidates. This is borne out in our research from the 2024 May polls, which showed a significant number of candidates in England reported experiencing harassment, intimidation and threats to their security (see paragraphs 15-24 below for more detail).

Our survey of public attitudes, conducted in February 2024, also showed bullying behaviours are more likely to be deemed acceptable if directed at a politician rather than a member of the public. Younger people were less likely to say bullying behaviour is unacceptable, regardless of the target being a member of the public or a politician. The disparity in attitudes was also significantly greater among young people. This is particularly concerning: if younger cohorts maintain these attitudes as they age, the problem of abuse of candidates can become worse.

Unacceptable levels of abuse and intimidation

After the general election, we asked candidates whether they had experienced any problem with harassment, intimidation, or threats. Our survey found that 70% of respondents said they experienced one form of abuse or harassment (social media abuse, physical abuse, threatening behaviour etc) at least once. A quarter of candidates (26%) said they experienced four different types of harassment, intimidation or abuse at least once. A third of respondents said they were intimidated or intentionally made to feel unsafe at least once during their campaign. The same proportion had their campaign assets torn down or destroyed, while 30% had their property or campaign assets defaced or damaged.

Our research also revealed a clear deterrent effect on democratic participation – the majority (56%) of respondents to our survey of candidates said they avoided some form of campaign activity at least once due to a fear of abuse. One third (32%) avoided discussing controversial topics to avoid harassment.

Candidates provided detailed insights about these deterrent effects in follow-up interviews: 

I would like much clearer rules around public discourse. Some of the things said about me, my team and my campaign by opposition campaigners have had a very negative effect on my mental health and seriously make me consider standing again at any level.

Female, England, 35-44; Received damage and destruction of campaign materials, social media abuse, and unwanted physical contact.

Candidates also highlighted concerns about abuse directed at their campaign volunteers:

What upset me in this election was the amount of abuse volunteers acting on behalf of candidates received. I saw older female volunteers being routinely shouted at and being called rude names by people who have professional jobs. It seemed to be open season on people if they were holding a political leaflet. It is one thing to have to expect abuse as a candidate, but we cannot allow volunteers to be insulted.

Female, England, 45-54; Received social media abuse, threats, intimidatory behaviour, physical abuse, and was belittled or insulted.

Disproportionate levels of threats towards female and ethnic minority candidates

Our survey of candidates also demonstrated concerning patterns of targeted abuse toward historically underrepresented groups in Parliament. Respondents from ethnic minority backgrounds (34%) and female candidates (20%) highlighted they were more likely to report having experienced serious abuse during the UKPGE.

Female candidates faced particularly concerning forms of abuse online, including reports in the media of 400 instances of deepfake pornography targeted at senior women politicians.

When choosing from the examples of the intimidation they experienced, female respondents to our survey were more likely than male candidates to report being touched, hugged, or kissed against their will. Female candidates were also more likely to report avoiding campaigning alone (66% compared to 44% of all respondents). One female candidate explained her decision to stop solo campaigning:

When I was going around knocking on doors, I had really abusive males. They were just full-on, in your face saying [my party] was a load of crap and politics was wrong, and they were swearing at you and, you know, it was really unpleasant. […] I just stopped knocking on the doors and just leafleted in the end. My husband did most of the door knocking for me instead because being a male he had a better rapport. But I still think that it's incredibly sad because they obviously don't respond to a female wanting to put themselves forward.

Female, England, 45-54; Received damage and destruction of campaign materials, social media abuse, and intimidatory behaviour.

This example illustrates the broader "ripple effect" of abuse, where some candidates feel that they need to rely on family members for protection during campaign activities. Such decisions suggest that political participation may create burdens not just for candidates, but for their entire support network.

Candidates from ethnic minority backgrounds also experienced distinct patterns of abuse, being more likely to receive offensive social media posts about their ethnicity (55%) or religion (41%). Ongoing international tensions in some cases led to an unacceptable levels of antisemitic and Islamophobic abuse during the campaign period.

The impact of this abuse created additional barriers for ethnic minority candidates' political participation. Insights from our survey of candidates reveal some candidates from ethnic minority backgrounds faced difficult choices about reporting racist abuse, citing both the overwhelming volume of incidents and concerns about it overshadowing their campaign messages:

I reported the most concerning one. The racists comments on the doorstep were too many to keep counteracting. And. I just wanted to campaign.

Female, England, 35-44; Received threats, social media abuse, damage and destruction of campaign materials, intimidatory behaviour, and was belittled or insulted.

As the only minority ethnic candidate, it would have harmed my campaign as it could have become part of discourse as opposed to focusing on the issues facing residents.

Male, England, 25-34; Received threats, social media abuse, and intimidatory behaviour.

Impacts of disinformation at the UKPGE

Another contributing factor to abuse and intimidation is the spread of disinformation about candidates on social media platforms, where it can quickly reach large audiences. Platform algorithms can rapidly amplify misleading content to its users, contributing to abuse and intimidation of candidates online.

Over half (55%) of candidates responding to our survey experienced some form of harassment or abuse, with online spaces being where most harassment occurred (65%). Of the candidates who had problems with harassment, 43% said it came from anonymous sources. Almost a quarter (23%) of candidates also told us they avoided using social media at least once due to fear of online abuse and harassment.

We heard how some candidates felt election-related mis- and disinformation that was spread online led to in-person abuse with one candidate telling us they were:

punched by a stranger in a public house due to false information posted on-line, which the newspapers published an article on when they knew it was false.

England, Male, 55-64; Experienced social media abuse, physical abuse and intimidatory behaviour.

The spread of synthetic media like deepfakes can enable false narratives, create confusion and distrust by manipulating audio and visual content to mispresent individuals’ words and actions. This was not widespread at this general election but there were reports coordinated groups on X were spreading deepfake videos of politicians. These clips were altered to mislead the public about what the targeted politicians had said, alongside comments to create the impression the videos were real.

Our post-election survey of the public also found that while half of people (49%) took action like fact-checking when encountering misleading content, only 12% reported it to platforms. This low reporting rate suggests more needs to be done to help people understand how and when to act against harmful content. Age appeared to influence response rates, with younger people more likely to act (57% of those aged 18 to 34, compared with 46% of those aged 55 to 64 and 48% of those aged 65 to 74).

Respondents to our survey of candidates also said that they felt social media platforms were not doing enough to respond to online abuse and intimidation. One respondent, when asked why they did not report harassment online, answered:

it was only on Twitter and there is no point. There is also no point in reporting it to Twitter because it never meets their thresholds for action anymore.

Female, Scotland, 35-44; Received social media abuse.

Addressing the issue of threats against candidates, MPs and elections is a significant challenge. With multiple agencies and bodies holding different areas of responsibility, coordination is essential but challenging.

We remain committed to working with partners across the electoral community to address these concerning threats. Only through coordinated efforts can we foster a safe environment for candidates, campaigners (including volunteers) and electoral administrators.

Strengthening democratic education

A key focus of our forthcoming five-year Corporate Plan, which we will submit to Parliament in March, is to expand our democratic education work. Young people will be central to this work given the UK Government’s commitment to extending the franchise to 16- and 17-year-olds.

Strengthening democratic education for young voters forms a crucial part of a comprehensive strategy needed to address abuse and intimidation in politics.

As our research shows, young people are less likely to view abusive behaviour towards politicians as unacceptable. As we outlined in our response to the Curriculum and Assessment review, expanding opportunity for direct engagement between students and elected officials helps humanise politicians and increases trust in democratic institutions.

Our successful partnership with Ofcom and Shout Out UK on the Dismiss campaign at the UKPGE, which reached over 2 million young people aged 18-24, also demonstrates how digital literacy education can be effectively integrated into wider democratic education efforts.

Strengthening enforcement and protective measures

Our research shows that only one in five (21%) respondents who experienced abuse, harassment or intimidation reported it to police. When issues were reported, candidates had mixed experiences with police responses.

Operation Bridger brought a coordinated approach across different police forces during the UKPGE; however, we received evidence of inconsistencies in how police forces respond to reported incidents of abuse and harassment during the UKPGE. We were told that some candidates received no contact from their designated Single Point of Contact (SPOC) officers, even in hotly contested constituencies where threats had been reported. When incidents were reported, the quality and speed of responses varied markedly between police forces, with some allegedly inappropriately dismissing harassment claims as 'freedom of expression' without expertise in the police force.

More could be done to improve support for candidates – they need to be able to receive clear and timely information and guidance about how to access support. This could involve requiring candidates to provide police forces with contact details to allow them to share essential information and make contact in an emergency.

The Royal Canadian Mounted Police (RCMP) offers a different model for candidate protection. Like the UK, they provide candidates with written personal safety information and best practice guidance and establish direct RCMP points of contact for candidates. A distinctive measure is that they differentiate between criminal threats warranting charges and "lawful but awful" behaviour. This includes behaviours which may not meet the threshold for criminal charges but where officers may engage directly with individuals to acknowledge awareness of their activities.

There should also be dedicated points of contact and stable, well-advertised funding for security support to ensure candidates and elected representatives can access resources when facing abuse and intimidation.

Police forces and prosecutors must continue to treat allegations and cases of election-related intimidation seriously. They must demonstrate that those committing offences against candidates and campaigners will face significant sanctions.

There are also opportunities to further strengthen protections for candidates and voters within the electoral process, including considering, with police forces and Returning Officers, whether to establish secure zones where campaign activity would not be allowed around specific risk-assessed polling stations or count venues.

Political parties must also play a part in strengthening deterrents including ensuring in their include membership rules that explicitly emphasise respect for other campaigners and enable them to take appropriate action to sanction members if they are found to have abused or harassed another campaigner (for example removal of membership or deselection as a candidate).

Strengthening protections against online abuse

Social media and online platforms should also do more to help develop improved screening tools for candidates’ profiles, to remove abusive content and identify perpetrators.

Social media platforms would need to develop more sophisticated detection mechanisms for identifying abuse targeting candidates, to implement this recommendation. We think this could be achieved by investing in technology-based solutions (including using AI) for rapid content moderation and establishing more effective systems for identifying and sanctioning perpetrators of online abuse.

While some social media and online platforms have adopted measures to protect candidates and campaigners, there is a clear need for standardised, rapid-response protocols across all major social media platforms.

There is also an important role for Ofcom to play in this debate, given its new responsibilities and duties under the Online Safety Act 2023 (OSA). We welcome the establishment of Ofcom’s Advisory Committee on Dis- and Misinformation. We will work with Ofcom as they develop their new duties to protect content of democratic importance given that this content may also include election-related mis and disinformation.

Another option is following a scheme the Scottish Parliament developed to protect elected representatives from online abuse. Following a comprehensive review of security provisions, the Scottish Parliament Corporate Body (SPCB), commissioned a pilot social media monitoring scheme in 2023. The service uses targeted keywords and location data to identify potentially abusive content directed at Members of the Scottish Parliament (MSPs).

The pilot has shown promising results, with the first successful conviction secured using evidence of threatening and racist language against a MSP, gathered through the monitoring scheme. The service, which began as a 12-month pilot with 38 MSPs, identified approximately 500 cases that potentially met criminal thresholds. We understand the scheme has been made permanent and the SPCB has hired a second analyst to support the scheme. This approach combines a technological solution with evidence gathering to assist enforcement action against perpetrators.

Access to platform data is also essential for understanding how social media companies implement their content moderation policies. The European Union's Digital Services Act (DSA) provides a potential model for the UK, requiring large platforms to share data with vetted researchers studying systemic risks to electoral processes. Adopting a similar framework in the UK would enable evidence-based development of measures to protect candidates from online abuse, while allowing researchers and civil society to analyse patterns of harmful content during elections.

We also recommend campaigners to label any AI-generated or modified election material during the election. Our post-election survey of the public found nearly a quarter of voters saw election-related deepfakes, while 18% were unsure if they had encountered synthetic content. While some platforms have AI labelling policies, greater adoption across all platforms would complement digital imprint rules and help voters more easily identify AI-modified campaign materials.

Our future work to strengthen electoral resilience against threats

We are committed to undertake initiatives to understand what is driving abuse and intimidation in elections, and to collectively develop effective responses to tackle these problems.

During Welcome to Your Vote Week this March, we are supporting young people to access trustworthy information to get informed about politics, democracy and elections, and take the first steps to get involved locally. We have partnered with Shout Out UK to produce a series of lesson plans, short activities and videos that can be delivered by schools, colleges and youth groups. These resources increase young people’s knowledge of mis and disinformation, understanding of how to report it and equips them with tools to think critically about information.

It is also essential there is a clear shared understanding of what constitutes unacceptable behaviour towards candidates and campaigners, as opposed to robust political debate.

We will conduct research with the public to develop a clearer understanding of where they consider the threshold lies between acceptable and unacceptable behaviour. This research will help us to develop public campaigns and partnership activities to raise awareness of the fact that abusive behaviour towards campaigners and elected officials is never acceptable. It will also help to inform the policing approach to candidate safety so that it reflects the wider public view about what is legitimate campaign discourse and what is unacceptable abuse.

We will also make sure that candidates and campaigners understand what is considered abuse and intimidation and know how to report it.