Estimated headline crime in England and Wales remained statistically unchanged in the year ending June 2026, while survey estimates of fraud and computer misuse rose and police-recorded offences involving knives or sharp instruments fell. The latest figures bring together two different measures: a household survey of people’s experiences and records of offences reported to or identified by police.
The Crime Survey for England and Wales estimated 9.7 million incidents of headline crime in the year to June. That was not a statistically significant change from the approximately 9.3 million incidents estimated for the year ending June 2025. The difference between the rounded totals should not be read as evidence that overall crime rose: the statistical assessment found no significant change.
Fraud and computer misuse increased
The survey estimated 4.5 million fraud incidents, 11% more than the 4.1 million estimated a year earlier. The increase was attributed largely to bank and credit account fraud, estimated at 2.8 million incidents after a 13% rise. Computer misuse incidents were estimated at 817,000, up 23% from 663,000.
These are estimates of people’s reported experiences, not a count of every incident known to authorities. The survey asks household residents aged 16 and over about crimes experienced in the preceding year. It does not cover every population or offence, including crimes against businesses and people living in communal establishments. Because it samples people rather than interviewing the whole population, its estimates carry uncertainty.
Fraud and computer misuse also require particular care when comparing police records over time. Many incidents are never reported, so recorded figures do not measure the full scale of these offences. A new national reporting system introduced in December 2025 replaced the previous reporting arrangements, and changes in how information is collected and classified may affect recorded year-on-year movements. The survey estimates and police-recorded totals therefore describe different things and should not be combined as if they were a single count.
Police records show fewer knife offences
Police recorded 48,615 offences involving knives or sharp instruments in the year ending June 2026, 6% fewer than the 51,567 recorded in the preceding year. These records cover offences reported to or identified by police and offer a different view from the survey, including offences and populations outside its scope.
Police-recorded crime can be useful for tracking offences that are more likely to come to police attention, including lower-volume serious offences. But changes in public reporting, police activity and recording practices can influence the totals. The figures are not a direct measure of every offence committed, and a fall in recorded offences should be described as a fall in police records rather than proof that all knife-related offending declined by the same amount.
Domestic abuse estimates carry a comparability warning
The bulletin also cautions that changes to the survey questionnaire and its implementation may have affected domestic-abuse estimates and the scale of the increase previously reported for the year ending March 2026. Those estimates are presented separately from headline crime, and the latest domestic-abuse figures cover the year ending March, not June. The statistical authority says one year of data is not enough to establish whether the higher estimate represents a new trend.
More broadly, domestic abuse, sexual victimisation, stalking and harassment are not part of the 9.7 million headline-crime total. They are measured separately because they use distinct questions and methods. The bulletin also notes that the latest sexual-victimisation questions were redeveloped, making the newest estimates not comparable with previously published measures.
The figures are for England and Wales, not the United Kingdom as a whole. The survey’s headline estimate covers selected offences against households and people aged 16 and over; police records cover a wider range of offences and populations, but remain affected by reporting and recording. Read together, the latest release points to a broadly stable survey measure of headline crime, higher survey estimates for fraud and computer misuse, and fewer police-recorded knife or sharp-instrument offences, with the limitations of each measure kept in view.
