How Worthing Became the 'Shoplifting Capital of the UK'

Worthing’s “shoplifting capital” label came from a real recorded-crime ranking. It did not establish that the town had more underlying shop theft than anywhere else in Britain. Making offences easier to report can increase the count. Dividing that count by the number of residents answers a different question from measuring risk to shops.

The useful question is what the ranking actually tells us. Worthing shows why police leaders need to explain how a number was produced before allowing it to become a judgement about a whole town.

Where the shoplifting-capital headline came from

On 17 February 2025, Radio Worthing reported an Accu Components study placing Worthing first at 23.2 recorded shoplifting incidents per 1,000 residents. The report attributed the underlying statistics to ONS, NISRA and the Scottish Government. It did not provide enough methodological detail to reproduce the entire cross-UK ranking from the article alone.

That last qualification is useful. A UK league table needs comparable periods, offence definitions and geographic areas across different recording systems. Naming official sources does not establish that the resulting comparisons are equivalent.

The coverage soon made a broader claim. These existing headline cards reproduce wording and dates from author-supplied captures of the publisher pages. They are redraws, with browser controls and advertising removed.

Those headlines establish how the story was framed. They cannot establish the prevalence of theft. TAP’s examination of police confidence and media coverage explains why measuring the coverage and measuring public opinion are separate tasks too.

Worthing had changed how theft reached the police

Sussex Police’s account of 22 April 2026 says One Touch Reporting began in Adur and Worthing in 2024 to make shop theft easier to report. The force attributes an increase in reports to that change and describes targeting repeat offenders through Operation Apprentice.

The same release quotes PCC Katy Bourne reporting a Worthing town-centre average of 28 recorded offences a month from March 2025 to February 2026, against 47 a month in the corresponding previous twelve months. These are a stakeholder’s published figures for a defined area and period, rather than an independent evaluation of the reporting scheme.

Better reporting is a credible explanation for part of the high recorded rate. The public material reviewed here does not tell us what proportion it explains. Nor does the later fall prove that the intervention caused it: that would require a credible comparison and attention to other changes over the same period.

This is not an argument for dismissing retailers’ experiences. A shop can experience repeated theft and frightened staff in a town whose recorded total is falling. Local harm and a statistical ranking are different questions.

Different tables answer different questions

Dividing thefts by resident population is convenient. But shops also serve commuters, visitors and people travelling in from neighbouring areas. A residential rate measures offences relative to those who live there; it is not a direct measure of risk per shop, customer visit or trading hour.

CrimeRate’s shoplifting table, labelled updated 7 June 2026 and checked for this review, uses daytime population for its town and city comparisons. Its displayed town and city table is headed by Kensington and Chelsea, Middlesbrough and Brighton and Hove; Worthing is not included. Its processing and data periods also differ from the earlier ranking. This is not a controlled comparison showing that changing the denominator alone changed the order.

Neither table supplies a definitive league of underlying shop theft. Each makes choices about geography, population and the data included. Readers need those choices before they can decide whether the comparison serves their question. The same checks are set out in How Police Leaders Can Read Data Better.

The latest national fall needs a recording caveat too

The ONS bulletin for the year ending March 2026, published on 23 July, reports 507,086 police-recorded shoplifting offences in England and Wales, down 4% from 530,324 a year earlier. These are national figures, not Worthing’s.

ONS also identifies a change that readers could easily miss. In April 2025, the Home Office clarified that a retail theft involving violence or threats at the time of the theft or escape should be recorded as robbery of business property. ONS says that clarification contributed to the rise in that robbery category and probably had a small effect on reducing recorded shoplifting. The bulletin does not quantify the effect in Worthing.

A lower number in one category can therefore partly reflect where an offence is counted. That does not make the fall meaningless. It means an assessment of retail harm should examine violence and robbery alongside shoplifting.

The ONS user guide explains the wider limit: shoplifting is not covered by the household Crime Survey for England and Wales, and police records miss unreported offences. The Home Office’s Commercial Victimisation Survey provides another source, but ONS cautions that its smaller sample can make annual estimates volatile. There is no single national series that settles Worthing’s position for us.

Explain the measurement before defending the number

Forces should publish the expected recording effect when they introduce easier reporting. That means saying what changed, when it changed and which additional measures will help distinguish improved reporting from increased offending. It does not mean predicting an increase that has not yet happened or treating every unwelcome rise as a recording issue.

When a ranking appears, force analysts and communications teams should put the data period, boundary and population measure beside their response. A comparison of a town-centre neighbourhood with a whole borough is not repaired by using the same town name. Neither should April’s account of a fall be presented indefinitely as the current position.

A useful retail-crime account would also report repeat victimisation, violence against staff and criminal-justice outcomes where the underlying data support comparison. Forces should say which measures are incomplete and which come from retailers rather than police systems. The fuel-theft analysis examines the related problem of counting offences without examining what happens to them.

These are proposals for clearer public explanation, not findings from a trial of police communications. Their test is practical: can a reader identify what changed, how it was measured and what remains unknown? The five habits for reading police statistics provide a short routine for applying that test.

Worthing does not need a competing claim that everything is fine. It needs an account of retail crime that takes shopworkers seriously and gives the statistics their proper limits. Leaders should be willing to publish bad news. They should be equally willing to challenge a comparison that cannot support its headline.

Updated 7 October 2026: refreshed the national evidence, added the robbery-recording qualification and clarified that different commercial rankings do not isolate the effect of their population measures. This is a targeted source review, not a new estimate of shoplifting in Worthing.