AKA: law & order, safety, police, policing, violence in society
Key notes
- There is no measure of all crime over all years. ‘Victimisations (police stations)’ started in July 2014 and is widely regarded as the best indicator.
- Days per month can vary from 28 to 31, so this impacts both the amount of crime and per capita crime in a month. All other things being equal, February months of 28 days should have 9.7% less crime than 31-day months. As such, we've applied an adjustment so that months can be compared fairly, apples with apples. From July 2014, when the data began, until the latest data point, the average number of days in a month is 30.448 (3dp), so we have adjusted every month's score in the graph above to this, e.g., 28/30.448 for February months (outside leap years with 29 days) and 31/30.448 for December months.
- Remember that each Labour-led or National-led government has had coalition partners that share the successes, and the failures, with them. A full list of governments for this time range is at the bottom of this page.
Data sources
Data shown:
- Victimisations per capita per month (weight adjusted by day, too).
- July 2014 onwards, which is when this data series started.
Full data:
Frequency:
Updated:
- 2-month delay for victimisation data (1-month for offender data) and published on the last working day of each month
Last update:
Next update:
Why this KPI?
- There is no single measure of all crime in New Zealand over time.
- The most reliable long-term indicator available is Victimisations (police stations), which began in July 2014 and is updated monthly (with a two-month delay).
- This dataset records each time a crime is reported to police and a victim file is opened.
- It focuses on person-centred crime in society.
- It captures the major “visible” offences that most people associate with crime, under these five ANZSOC (Australian and New Zealand Society of Criminology) Divisions:
- Acts Intended to Cause Injury (Assault)
- Sexual Assault and Related Offences
- Abduction, Harassment and Other Related Offences Against a Person
- Robbery, Extortion and Related Offences
- Theft and Related Offences
- The NZ Police make the “Note: Burglary victimisations are classified to property rather than the resident and have been excluded from the data”.
- Several ANZSOC Divisions are excluded from the victimisation data, often due to very small sample sizes, e.g., murders, no 3rd-party victims, e.g., speeding tickets, or victim counts being hard to identify, e.g., fraud:
- Homicide and Related Offences
- Dangerous or Negligent Acts Endangering Persons
- Fraud, Deception and Related Offences
- Miscellaneous Offences
- Offences Against Justice Procedures, Govt Sec and Gov Ops
- Property Damage and Environmental Pollution
- Public Order Offences
- Traffic and Vehicle Regulatory Offences
- Prohibited and Regulated Weapons and Explosive Devices [unlikely to have victims]
- Two important caveats from the NZ Police:
- ‘The Ministry of Justice survey shows that only about 25% [of crime] is reported and with varying degree across different crime types.”
- ‘The 'Victimisations (police stations)' report has the complete dataset from July 2014.’
- Proceedings data can't be used as it only record crimes where an offender is identified and processed. That’s a much smaller number, e.g., June 2025 Victimisations (police stations) was 27,062, whereas Proceedings (police stations) was only 8,737 (32% as many). Proceedings are also impacted by policy changes to deprioritise certain crimes, e.g., shoplifting could be increasing while shoplifting proceedings decrease as resources are focused elsewhere.
- All other reports start too late to be as useful, e.g., 2019-2020, and also don't count all crimes.
Related facts
Wishlist
With support, we’ll be able to add multiple KPIs for each issue:
- Updates on many of KPIs listed above.
- NGL (National Gang List) total (older data here)
- Murder convictions (older data here)
- Police staff per capita (frontline vs. back office)
- Domestic violence
- Child abuse
- Prisoners per capita (needs to be paired with crime rates)
- Cases > Arrests > Charges > Convictions (per capita and then %s)
- Average sentences (by offence class)
- Reoffending
- Retail crime (hard to measure)
- Youth crime
Discarded
- Offender-based crime statistics. Offences are only measured if the offender has been caught. As such, offender statistics are more useful for working out crime rates and types by demographic.
Governments over the timeframe shown
- 2008 = National + ACT, United, Māori
2011 = National + ACT, United, Māori
2014 = National + ACT, United, Māori
- 2017 = Labour/NZ First + Green
2020 = Labour + Green
- 2023 = National/ACT/NZ First
Sources:1,2 (table 5).
Data published by NZ Police
© Crown Copyright
Licensed for use under the creative commons attribution licence (BY) 4.0
If you spot a typo, mistake, or improvement opportunity for this page, please let us know via the comments below or contact us!
