Hate crime statistics for England and Wales, year ending March 2026

The police record hate crime against five nationally monitored strands. Race is by far the largest: 107,926 offences, about 74% of all recorded hate crimes. Race and religion rose over the year; the other three strands fell.

What moved within each strand

Race. The Home Office says the rise in race hate crime was due to an increase in public order offences (up 10%, or 5,490 offences) and a 10% rise in assault without injury (up 1,460 offences). Leaving out the Metropolitan Police, the 90,378 race hate crimes recorded in 2025/26 were the highest annual total since the data was first collected in the year ending March 2012 (bulletin section 2.2).

Religion. Excluding the Met, the 8,214 religious hate crimes were also the highest annual total since records began in the year ending March 2012. There is more on religion below.

Sexual orientation. Excluding the Met, these offences have fallen 17% since the year ending March 2022, from 22,317 to 18,520. The Home Office puts this partly down to fewer malicious communications offences after a 2023 change to the counting rules for conduct crimes such as stalking and harassment, which removed the requirement to record 2 crimes when one of them was a conduct crime.

Disability and transgender. Excluding the Met, both fell for the third year running: disability from 13,637 in the year ending March 2023 to 9,616 (down 29%), and transgender from 4,477 to 3,643 (down 19%). The bulletin records falls in harassment, malicious communications and assault for both.

This is the question the figures are most often used to answer, and the Home Office is careful about it. Police recorded hate crime measures what the police record. That depends on what people report, how forces record crime and how consistently they flag hate crime, and the Home Office says recording and the identification of hate crime have both improved a great deal since 2014.

To compare years fairly, the Home Office leaves out the Metropolitan Police, whose figures changed when it moved to a new recording system in February 2024 (see how to read these numbers). On that basis, recorded hate crime rose from 31,884 offences in 2012/13 to a peak of 128,485 in 2021/22, fell for 2 years to 113,166 in 2023/24, and has risen in each of the last 2 years, to 123,957 in 2025/26 (up 7% on 2024/25, our calculation from Home Office figures).

Why the long-term rise is mostly about recording

In 2014 an HMICFRS inspection concluded that 33% of cases involving violence were not being recorded (bulletin section 4.4). HMICFRS's August 2025 report, cited in the bulletin, estimates the share of crimes correctly recorded by the police rose from 80.5% in 2014 to 94.8% for 2023 to 2025, and for violent crime from 66.9% to 93.6%. The Home Office thinks recording of public order offences improved in a similar way, and those two offence groups make up 93% of recorded hate crime. It adds that growing awareness of what counts as hate crime has improved how officers identify it. For these reasons it says police recorded crime figures do not provide reliable trends in hate crime since 2014 (bulletin section 2.1).

The Home Office's test is to compare hate crime with all violence and public order offences, indexed to the same starting point. Up to 2021/22 the two lines move closely together, which the Home Office says suggests improvements in crime recording were an important factor in the increase. Since then they have separated: in 2025/26 the hate crime index stood at 321 against 285 for violence and public order offences (2014/15 = 100). In the bulletin's words, the rise over the last 2 years "may suggest there has been a genuine increase in hate crime offences".

Month by month: racially or religiously aggravated offences

Five offences in law are specifically racially or religiously aggravated, such as racially or religiously aggravated harassment. They are hate crimes by definition, and the bulletin says 58% of hate crimes were recorded as one of them (section 1.4). Unlike hate crime as a whole, they are reported monthly for every force, so short-term changes can be seen. In the 12 months to March 2026 the police recorded 92,725 of these offences (our total of the monthly figures).

Police figures are not the same as how much hate crime happens

Not all hate crime is reported to the police, so the Home Office says police figures "should also not be seen as a good measure of prevalence". They are, in its words, a good measure of the hate crime-related demand on the police (bulletin section 1.1).

The other source is the Crime Survey for England and Wales, run by the Office for National Statistics. It asks people about crimes they experienced, whether or not they told the police, and is unaffected by changes in police recording. But a single year's sample is too small for hate crime, so three years are combined, and it does not cover public order offences or crimes against businesses. The latest survey estimates covered April 2022 to March 2025 and were published last year; the next are due in 2028 (bulletin sections 1.6 and 4.4).

Police recorded 11,419 religious hate crimes, up 13%. The police record the religion the offender targeted, which is usually but not always the victim's own. Hate crimes targeted at Muslims rose 15% to 5,132 (46% of religious hate crimes), and those targeted at Jewish people rose 10% to 3,162 (28%).

Excluding the Met, offences targeted at Muslims (3,660) were at their highest annual level since this data was first collected in the year ending March 2018. Offences targeted at Jewish people (2,003) were at their second highest, slightly below the 2,093 recorded in 2023/24 (bulletin section 2.2; data tables, table 5a).

Rates per 10,000 people

Counts alone favour larger groups, so the Home Office also gives rates using 2021 Census population figures. Police recorded 117 religious hate crimes per 10,000 Jewish people, the highest rate for any religious group, and 13 per 10,000 Muslims. The Crime Survey, which asks people directly, shows a different pattern: it estimated that 0.8% of Muslims and 0.4% of Jewish people experienced a religious hate crime per year on average (April 2022 to March 2025). The Home Office says the difference may be due to how likely a crime is to be reported to the police and the different coverage of the two sources.

What kind of offences

As in previous years, public order offences, specifically causing public fear, alarm or distress, were the most common type of religious hate crime against both groups. Criminal damage made up a larger share of offences targeted at Jewish people; assaults made up a larger share of those targeted at Muslims.

Hate crime looks very different from crime as a whole. Just over half (53%) of recorded hate crimes were public order offences, mostly causing public fear, alarm or distress, against 9% of all police recorded crime. Theft, which is almost a third of all recorded crime, is less likely to involve a motivating factor against a monitored strand (bulletin section 2.3).

Public order offences were the most common type for every strand except disability, where stalking and harassment came first (41.2% of disability hate crimes).

10% of hate crime flagged offences had been dealt with by a charge or summons. The Home Office notes this was higher than the previous year's 8%, but that figure came from 35 forces, so the two are not directly comparable. Because hate crime is mostly public order and violence, the fairest comparison is offence group by offence group.

For violence against the person the rates were the same once rounded (7% and 7%), and the Home Office says they have moved closer in recent years. For criminal damage and arson they were also the same (5%). Hate crime flagged public order offences were more likely to end in a charge or summons (12% against 9%).

Racially or religiously aggravated offences

Outcome data for the five aggravated offences is available for every force. Aggravated public order and assault offences were more likely to be dealt with by a charge or summons than their ordinary equivalents, which the Home Office says reflects the more serious nature of the aggravated offences. For criminal damage it was the other way round. Across all aggravated offences the charge or summons rate rose from 8% in the year ending March 2024 to 10%, reversing a long downward trend: in the year ending March 2015, 30% of aggravated public fear, alarm or distress offences ended in a charge or summons (bulletin section 2.5).

How long it takes

Hate crimes took longer to reach a final outcome. The median for criminal damage and arson was 18 days against 6 for other offences, and for violence against the person 35 days against 20. The Home Office says this may be due to the complexity of these crimes, or the greater investigative effort invested.

The full picture of outcomes

Most hate crime flagged offences did not end in a charge. The two largest outcomes were investigation complete with no suspect identified (29.9%) and evidential difficulties where the victim does not support further action (28.3%). A further 17.5% closed with evidential difficulties although a suspect was identified and the victim supported action. For violent offences, evidential difficulties of either kind accounted for 51% of hate crimes and 61% of other violent offences (bulletin section 2.5; appendix table 5).

7% of hate crimes (8,028 offences) were flagged as having an online element, a similar proportion to all police recorded crime (6%). The online share was highest for transgender hate crime and lowest for race. Most online hate crimes were stalking and harassment (43%) or malicious communications (39%). When the Home Office last published online figures, for the year ending March 2018, 2% of hate crimes were identified as online, but it stopped then because of concerns about accuracy.

Where the victim's self-defined ethnicity was known, 34% identified as Asian or Asian British, 29% as White and 24% as Black, Black British, Caribbean or African. The Home Office says offences against White victims are likely to include xenophobic abuse against people not born in the UK.

The table covers all 43 territorial forces and the British Transport Police. Together they recorded 146,825 hate crimes. 32 forces recorded more than in 2024/25 and 12 recorded fewer (our calculation). The 2024/25 figures are as published in last year's release, where the England and Wales total was 137,550; this year's bulletin gives 137,560 for 2024/25, so small revisions are possible.

By region

The ten regions add up to 140,657, the total for the 43 territorial forces. The British Transport Police covers the railways across England and Wales and is not part of any region.

We read the numbers directly from the Home Office's official data tables and appendix tables with a script on 8 October 2026, the day they were published, so nothing was typed in by hand. Figures quoted from the bulletin text, such as the 58% aggravated share and the HMICFRS recording rates, were checked against the bulletin and are cited to its sections. Where we have worked something out ourselves, such as a force's change or its share of the total, the page says "our calculation". Percentages in tables are to one decimal place; in the text we use whole numbers, as the Home Office does.

We do not rank forces, add explanations the Home Office has not given, or treat recorded crime as a measure of how much hate crime happens. If you spot an error, our corrections policy explains how to tell us and what we do.