This article is intended to provide information on long-term trends alongside additional data on the characteristics of victims and nature of crime. It may not include the most recent published data, which can be found in the latest quarterly Crime in England and Wales release.Back to table of contents
A person commits robbery if in the course of stealing or attempting to steal from someone, they use force or the threat of force. This is sometimes colloquially referred to as “mugging”.
A person commits “theft from the person” if they steal property while the property is being held or carried by the victim. This includes snatch thefts (where some degree of force is directed to the property but not to the victim) and stealth thefts (where the victim is unaware of the offence being committed, for example, pick-pocketing). Unlike robbery, these offences do not involve violence or threats to the victim.
There can sometimes be a subtle distinction between robbery and theft from the person (if, for example, minimal force is directed toward the victim during the course of a robbery) and, as such, these offences do share common drivers, given the principal aim with both is to acquire property.
These offences are included in the Crime Survey for England and Wales (CSEW) and police recorded crime data, although the information available from each source and their coverage differ.
Police recorded crime covers both robbery of personal property and business property and these are separately identifiable in the published data. Data are also available on the numbers of robbery offences involving the use of a knife or sharp instrument.
The CSEW is a survey of the population resident in households and, as such, information collected on robbery offences only relate to personal property and not of business property. The only exception to this is if the business is the respondent’s business and they are the sole trader; otherwise, the incident would be treated as either an assault or a threat, depending on whether the offender used or threatened to use force.
Being a relatively low-volume offence, despite the large CSEW sample, the number of respondents who have been victims in the past year can be quite small. This means that although the CSEW provides a good indication of long-term and national trends, estimates tend to fluctuate from year to year and the police recorded crime series may provide a better measure of short-term and subnational trends.
Theft from the person
“Theft from the person” does not include incidents where force or threats of force (other than where minimal force is applied to snatch property) are used at the time the theft occurs; such incidents would be classed as robberies.
Within the CSEW, it is possible to separately identify three sub categories of “theft from the person”:
“snatch theft from person”, which occurs when property that is being held or carried by someone is stolen, but only minimal or no force is used to snatch the property away
“stealth theft from person”, which occurs when property that is being held or carried by someone is stolen, but at the time the theft takes place, the person is unaware of the theft (and no force is used)
“attempted snatch or stealth theft from person”, which includes unsuccessful attempts to steal property that someone is holding or carrying on their person
These sub categories form part of the typology developed for the CSEW and are not separate criminal offences, but provide some additional information about the manner in which the incident occurred.
Within the police recorded crime series, only a total for “theft from the person” is available.
The CSEW provides a better measure of national trends in “theft from the person”, especially for the sub-types of snatch theft and stealth theft, that are not separately identifiable in the police recorded crime data. But, police recorded crime data are available for total theft from the person offences at subnational level, where the sample size of the CSEW is not large enough to yield estimates with reasonable levels of precision.
Further discussion on the strengths and limitations of the two main sources is available in the “Which source provides the better measure of these offences?” section.Back to table of contents
The Crime Survey for England and Wales (CSEW) has collected information on crimes, including robbery and theft from the person, experienced by respondents in a consistent manner since the survey first ran in 1981 and thus estimates are directly comparable across the history of the survey.
Care should be taken with regard to historical police recorded crime data, as changes to recording practices following the introduction of the new Home Office Counting Rules (HOCR) in April 1998 and the National Crime Recording Standard (NCRS) in April 2002 typically resulted in an increase in the number of crimes recorded.
The long-term trends in robbery as measured by the CSEW and police recorded crime are broadly similar.
Robbery is a relatively low-volume offence, typically accounting for between 1% and 3% of crimes covered by the CSEW1 and less than 2% of all police recorded crime in England and Wales each year.
The estimated number of CSEW robbery incidents rose sharply through much of the 1990s (peaking at 407,0002 in the year to December 1999); this increase is consistent with the general increase in crime during this period. From then onwards, there has been a general pattern of decline (reaching a low of 89,0003 incidents in the survey year ending March 2015), albeit with some substantial year-on-year fluctuation.
In the survey year ending March 2017, around 2 in 1,000 adults had been victims of robbery (equivalent to less than half of 1%). This compares with around 7 in 1,000 adults in the year ending December 1997, meaning that the likelihood of being a victim of robbery fell by almost 70% compared with 1997.
The numbers of police recorded robberies rose steadily throughout the 1980s and up until the mid-1990s, before rising sharply year-on-year between 1999 and 2002. Following the introduction of the NCRS in April 2002, with the exception of a notable (11%) rise in robberies between the year ending March 2005 and the year ending March 2007, there has been a general downward trend between the year ending March 2003 and the year ending March 2016. There has, however, been an increase in the number of robberies recorded by the police in the year ending March 2017, though it is not clear at this stage whether this represents a change in the longer-term downward trend.
Since the year ending March 1999, when information for the separate offence classifications was first available, around 9 in 10 robberies recorded by the police were of personal property and around 1 in 10 of business property.
Robbery is one of the seven offences recorded by the police for which additional information is collected from the police regarding the use of a knife or sharp instrument during the course of the offence. Since comparable data are first available (from April 2010 onwards), there has been a drop of around 3,500 robberies recorded by the police that involved a knife or sharp instrument between the year ending March 2011 and the year ending March 2017, although there was an increase of around 2,500 offences between the year ending March 2016 and the year ending March 2017. This drop has mirrored the fall in all robberies recorded by the police over the same period and hence the proportion of all robberies that have involved a knife or other sharp instrument has remained consistent, at around 1 in 5.
Theft from the person
As with robbery, theft from the person is a relatively low-volume offence, typically accounting for between 3% and 7% of crimes covered by the CSEW4 and less than 3% of all police recorded crime in England and Wales each year.
The trends in theft from the person as measured by the CSEW and police recorded crime are broadly similar. However, the substantial year-on-year fluctuation in the CSEW estimates over time, owing to the low-volume of these offences, makes short-term trends in CSEW theft from the person more difficult to ascertain.
The estimated numbers of CSEW theft from the person incidents rose sharply through the early 1990s and generally declined until the late 2000s, albeit with some substantial year-to-year fluctuation. Since then, while there have been some year-on-year increases, there has been a general decline in the number of theft from the person incidents (down to 367,0005 in the survey year ending March 2017).
Stealth theft from the person comprises the majority of all CSEW theft from the person incidents, typically accounting for between 50% and 60% of all incidents (although there are exceptions in a few survey years). In contrast, snatch theft from the person typically only comprises between 10% and 20% of all theft from the person incidents, with attempted snatch and stealth theft from the person typically accounting for between the remaining 20% and 40% of incidents.
Consequently, the sub category of “stealth theft from person” has followed a similar trend to that of “theft from the person” overall. The sub category of “attempted snatch or stealth theft from person” has also followed a similar trend to “theft from the person” overall, albeit with less year-on-year fluctuation; whereas the trend in “snatch theft from the person” has remained relatively flat over the whole series.
Although typically around two to three times as prevalent as robbery, theft from the person is still a relatively low-volume offence. In the survey year ending March 2017, less than 1% of adults were victims of theft from the person; around 7 in 1,000 adults had been victims. This compares with around 16 in 1,000 adults in the year ending December 1995, meaning that adults are currently around half as likely to be a victim of theft from the person as in 1995.
The trend in police recorded theft from the person offences was relatively flat throughout the 1980s (between around 30,000 to 35,000 offences), gradually increasing through to the mid-1990s and more sharply increasing between the year ending December 1999 and the year ending March 2003 (from just under 60,000 offences to almost 150,000 offences). Since the year ending March 2003, the trend has generally been one of decline (falling to a low of just under 80,000 offences in the year ending March 2015). There was, however, a period between the year ending March 2009 and the year ending March 2013 where consecutive year-on-year increases were seen; this was also seen more recently between the year ending March 2015 and the year ending March 2017.
The increases in these offences recorded by the police between the year ending March 2009 and the year ending March 2013 are thought to have been driven by thefts of smartphones. The subsequent decreases may, in part, be explained by improvements to mobile phone security and theft prevention (the Home Office report Reducing mobile phone theft and improving security published in September 2014 has more information).
CSEW data on robbery and theft from the person are also available for children aged 10 to 15. In the survey year ending March 2017 there were an estimated 41,0006 incidents of robbery and 36,0007 incidents of theft from the person experienced by children aged 10 to 15. Given the small sample size for the 10- to 15-year-old element of the CSEW, estimates for both these offences can greatly fluctuate over time and as a result trends can be difficult to interpret. Detailed data are published in Appendix tables A9, A10, A11 and A12 alongside the quarterly Crime in England and Wales releases.
Latest figures and trends in robbery and theft from the person offences are available in the most recent quarterly Crime in England and Wales release.
Notes for: What are the long-term trends?
Excluding fraud and computer misuse; the CSEW only started measuring these offences in October 2015. Robbery incidents accounted for 1% of CSEW crime (including fraud and computer misuse) in the survey year ending March 2017.
The 95% confidence interval around this estimate is [269,000 to 544,000]. Any sample survey may produce estimates that differ from the figures that would have been obtained if the whole population had been interviewed. The confidence interval provides a range of values around an estimate (also referred to as the margin of error of the estimate). Section 8.1 of the User Guide to Crime Statistics for England and Wales has more information about confidence intervals and statistical significance.
The 95% confidence interval around this estimate is [61,000 to 116,000].
Excluding fraud and computer misuse; theft from the person incidents accounted for 3% of CSEW crime (including fraud and computer misuse) in the survey year ending March 2017.
The 95% confidence interval around this estimate is [312,000 to 422,000].
The 95% confidence interval around this estimate is [14,000 to 68,000].
The 95% confidence interval around this estimate is [19,000 to 52,000].
Robbery is an offence that is geographically concentrated in a few large urban areas. For example, it is fairly consistent that around 60% of robberies recorded by the police in the whole of England and Wales are recorded in just three force areas: the Metropolitan Police (whose area typically accounts for between around 40% and 50% of all robbery offences), West Midlands and Greater Manchester.
Owing to robbery being a low-volume offence in the Crime Survey for England and Wales (CSEW), there is less variation around prevalence estimates and hence it is not possible to identify many demographic groups that are consistently more likely to be victims of robbery, however:
- males tend to be more likely victims than females
- younger adults (aged 16 to 34) tend to be more likely victims than older adults (35 and over)
Theft from the person
As with robbery, theft from the person offences are concentrated in a few predominantly urban areas; although unlike robbery, the British Transport Police also record a large number of incidents, reflecting the volume of these offences that occur on the railway network. The Metropolitan Police typically accounted for around 40% of all recorded theft from the person offences in recent years. They, along with Greater Manchester and West Yorkshire police forces and the British Transport Police, consistently accounted for between 50% and 60% of recorded theft from the person offences.
Owing to theft from the person being a low-volume offence in the CSEW, there is less variation around prevalence estimates and hence it is not possible to identify many demographic groups that are consistently more likely to be victims of theft from the person, however:
- in contrast to robbery, females tend to be more likely victims than males
- mirroring robbery, younger adults (aged 16 to 34) tend to be more likely victims than older adults (35 and over)
The Crime Survey for England and Wales (CSEW) Nature of Crime tables provide further information about the circumstances surrounding incidents of robbery over the last decade, including:
- the large majority of robberies were committed by men (between 84% and 97% in each survey year)
- the age profile of robbery offenders has been increasing; in the year ending March 2006 survey the offender(s) were aged 25 or older in around one-fifth of robberies compared with over half of robberies in the latest survey year- this is supported by Ministry of Justice sentencing data, which indicate falls in the numbers of younger people (aged under 21) convicted of committing robbery
- cash and foreign currency have typically been the most frequently stolen items in robberies, with mobile phones also being commonly stolen
- weapons were used in between around one-fifth and one-third of robberies; and where a weapon was used, it was most commonly a knife
Theft from the person
The Nature of Crime tables provide further information about the circumstances surrounding incidents of CSEW theft from the person. Findings over the last decade include:
- purses, wallets and money have typically been the most frequently stolen items in thefts from the person, with mobile phones also being commonly stolen
- the most common location where thefts from the person occur was typically the street, although inside shops and public transport were also common locations (the latter reflected in the comparatively high number of offences recorded by the British Transport Police)
- between three-fifths and two-thirds of thefts from the person took place during the week (equivalent to around 14% per weekday) and between one-third and two-fifths took place during the weekend (equivalent to 15% per weekend day)
- typically around two-fifths of thefts from the person took place during the afternoon (noon to 6pm) and around one-quarter during the evening (6pm to midnight)
Further information about the nature of incidents is currently restricted to the CSEW; only limited data are currently available on the circumstances surrounding offences in the main recorded crime collection. A special Home Office data collection on the use of knives or sharp instruments for selected offences, including robbery, is available and additional information on this can be found in Focus On Violent Crime and Sexual Offences. It is anticipated that in the future we may be able to provide further information about crime incidents recorded by the police as more detailed data sources become available centrally.Back to table of contents
The Crime Survey for England and Wales (CSEW) has collected information on crimes, including robbery and theft from the person, experienced by respondents in a consistent manner since the survey first ran in 1981. It captures incidents that are not reported to the police and is not affected by changes in police recording practices; it is therefore a more reliable measure of long-term trends than police recorded crime.
However, the survey will not capture crimes against businesses, so data on robbery of business property are not available, or offences committed against people not resident in households (for example, students in halls of residence or overseas visitors). While estimates at the national level (England and Wales) are of good quality, lower-level geography estimates are not robust.
Additionally, with both robbery and theft from the person being low-volume offences in the CSEW, estimates are prone to greater fluctuation than estimates for other, more frequently occurring, offence types. Thus, police recorded crime can often be a better guide to short-term trends in robbery and theft from the person than the CSEW.
Police recorded crime data have a wider population coverage, including crimes committed against people not resident in households (such as overseas visitors) if reported to and recorded by the police. Lower-level geography data (police force and community safety partnership areas) are also available.
Additionally, police recorded robbery data have a wider offence coverage – covering business as well as personal property.
However, the police recorded crime data do not include offences that do not come to the attention of the police or are not recorded by them. According to the CSEW, robbery is one of the offences more commonly reported to the police (between an estimated two-fifths and three-fifths of victims do so), reflecting the severity of the offence. But theft from the person tends to be reported to the police slightly less frequently; in most survey years, snatch thefts and stealth thefts from the person are reported by between an estimated two-fifths and half of victims.
Also, due to changes in recording practices introduced in 1998 and 2002, it is not possible to directly compare police recorded crime data for any period prior to the year ending March 2003 with subsequent years.
Additionally, there are currently concerns about the quality of crime recording; crimes may not be recorded consistently across police forces and so the true level of recorded crime may be understated. Following an assessment of crime statistics by the UK Statistics Authority, published in January 2014, the statistics based on police recorded crime data have been found not to meet the required standard for designation as National Statistics.
Since the UK Statistics Authority assessment decision, Her Majesty’s Inspectorate of Constabulary (HMIC) has undertaken an inspection of the integrity of police recorded crime. The Crime-recording: making the victim count report published by HMIC found that an estimated 14% of robbery offences that should have been recorded as a crime were not. However, this level of under-recording was lower than for all offences on average (19%).
The renewed focus on the quality of crime recording means that caution is needed when interpreting statistics on police recorded crime. While it is likely that improvements in compliance with the National Crime Recording Standard (NCRS) may have led to increases in the number of crimes recorded by the police, it is not possible to quantify the scale of this, or assess how this effect varied between different police forces. However, the effect of improved compliance with recording standards is thought to have been more pronounced for violent and sexual offences rather than robbery and theft offences.
Therefore, on balance, the CSEW provides a better measure of national and long-term trends in robbery and theft from the person; although police recorded crime provides a better source for short-term trends and subnational data for these offences.Back to table of contents
Crime Survey for England and Wales
Crime in England and Wales (quarterly publication) – the preferred source for latest trends:
- Appendix tables A1, A2, A3 and A8 include data on numbers of incidents, incidence rates, prevalence rates and numbers of victims for the complete survey time series (starting from the year ending December 1981)
- Quarterly table QT2 includes data on numbers of incidents in the previous 3 survey years, broken down by quarter of interview
- Open data tables (Personal Crime – Incidence and Personal Crime – Prevalence) include data broken down by demographic information
- Annual trend and demographic tables D5, D6, D7 and D8 include data on repeat victimisation and incidents reported to the police – note: only published alongside “Year ending March” releases
- Annual supplementary table S14 includes data on victim satisfaction with police handling of theft from the person incidents – note: only published alongside “Year ending March” releases
Focus on: Property Crime (annual publication; latest edition published in November 2016) – the preferred source for more detailed analysis, including victim characteristics and details regarding the circumstances of offences:
- commentary in overview chapter on time-series trends and more detailed findings from the latest financial year
- Appendix tables 6, 7, 8 and 9 include data on victim demographics
- Nature of crime tables 9.1 to 9.10 include data about the circumstances surrounding robbery incidents, including time, location, injuries sustained, weapons used and more
- Nature of crime tables 7.1 to 7.6 include data about the circumstances surrounding the theft from the person incidents, including time, location, item(s) stolen and more
Police recorded crime
Crime in England and Wales (quarterly publication):
- Appendix table A4 includes data on numbers of incidents since the year ending March 2003, following the introduction of the National Crime Recording Standard (NCRS) in April 2002; this is the earliest time period for which the data are directly comparable
- Bulletin table 3 includes data on the numbers and proportions of robbery incidents since the year ending March 2011 that involved the use of a knife or sharp instrument
- Quarterly table QT1 includes data on numbers of incidents in the previous 2 years, broken down by quarter
- Police force area tables P1, P2 and P3 include data on numbers of incidents and rates per 1,000 population in the latest 12-month period and percentage changes with the previous 12-month period, broken down at police force area level
- Police force area open data include a time series of numbers of incidents back to the year ending March 2003 by police force area
- Community safety partnership data include a time series of numbers of incidents back to the year ending March 2003 by community safety partnership
Focus on: Property Crime (annual publication; latest edition published in November 2016):
- commentary in overview chapter on time-series trends and more detailed findings from the latest financial year
Crime outcomes (Home Office)
The Home Office publishes data on the outcomes of crimes recorded by the 43 territorial police forces in England and Wales, plus the British Transport Police. The latest publication is available from the Home Office Crime outcomes in England and Wales statistics web pages.
Commercial Victimisation Survey (Home Office)
The Commercial Victimisation Survey (CVS) is a telephone survey in which respondents from a representative sample of business premises in certain sectors in England and Wales are asked about crimes experienced at their premises in the 12 months prior to interview. In 2016, for example, three sectors were surveyed: “Wholesale and retail”, “Transportation and storage” and “Administration and support”.
The CVS is run by the Home Office and data are published on the Home Office’s Crime against businesses statistics web pages.
Data are available on the proportion and number of premises that experienced robbery, as well as the incidence rates and number of incidents of robbery in the year prior to interview, by industry sector.
Although some data on thefts are collected by the CVS (for example, thefts by customers or employees), being a survey of business premises, thefts from the person are not covered.
Sentencing data (Ministry of Justice)
The Ministry of Justice (MoJ) publish data on prosecutions and sentencing; the latest Criminal Justice Statistics Quarterly (CJSQ) publication is available from the MoJ Criminal justice statistics web pages.
However, while robbery data are readily available, in most of their headline data, it is not possible to separately identify theft from the person offences from within “Theft offences”.
MoJ publish data tools on prosecutions and sentencing at an offence level breakdown (so thefts from the person are separately identifiable); these are available in the year ending December CJSQ releases.
Crime statistics for Scotland are collected and published separately.
Recorded crime statistics for Scotland are not directly comparable with those in England and Wales. The recorded crime statistics for Scotland are collected on the basis of the Scottish Crime Recording Standard, introduced in 2004, which like its counterpart in England and Wales, aims to give consistency in crime recording. The main principles of the Scottish Crime Recording Standard are similar to the National Crime Recording Standard for England and Wales with regard to when a crime should be recorded; however, there are differences between the respective counting rules.
Differences in legislation and common law also have to be taken into account when comparing the crime statistics for Scotland with England and Wales.
The Scottish Crime and Justice Survey (SCJS), however, does follow a similar format to the Crime Survey for England and Wales, having a shared antecedence in the British Crime Survey (whose sample during some rounds of the survey in the 1980s covered Scotland, south of the Caledonian Canal). So, while there are differences in the crimes or offence classifications to reflect the differing legal systems, the data are broadly comparable.
Police recorded crime and SCJS data are published by the Scottish Government.
Crime statistics for Northern Ireland are collected and published separately.
The legal system in Northern Ireland is based on that of England and Wales; the Police Service for Northern Ireland (PSNI) also has the same notifiable offence list for recorded crime as used in England and Wales. In addition, the PSNI has adopted the National Crime Recording Standard (NCRS) and Home Office Counting Rules for recorded crime that applies in England and Wales. Therefore, there is broad comparability between the recorded crime statistics in Northern Ireland and England and Wales.
The Northern Ireland Crime Survey (NICS) also closely mirrors the format and content of the Crime Survey for England and Wales, using a very similar methodology with continuous interviewing and a face-to-face interview with a nationally representative sample of adults (16 years and over), using a similar set of questions. Therefore, results from the two surveys are broadly comparable.Back to table of contents
The basic definition of theft is laid out in section 1 of the Theft Act 1968:
 A person is guilty of theft if he dishonestly appropriates property belonging to another with the intention of permanently depriving the other of it; and “thief” and “steal” shall be construed accordingly.
 It is immaterial whether the appropriation is made with a view to gain, or is made for the thief’s own benefit.
The offence of theft (which includes the sub-category of “theft from the person” as defined in the Home Office Counting Rules) is laid out in section 7 of the Theft Act 1968:
 A person guilty of theft shall on conviction on indictment be liable to imprisonment for a term not exceeding 7 years.
The offence of robbery is laid out in section 8 of the Theft Act 1968:
 A person is guilty of robbery if he steals and immediately before or at the time of doing so and in order to do so, he uses force on any person or puts or seeks to put any person in fear of being then and there subject to force.
 A person guilty of robbery, or of an assault with intent to rob, shall on conviction on indictment be liable to imprisonment for life.Back to table of contents
Contact details for this Article
Telephone: +44 (0)20 7592 8695