Exclusive:Watchdog calls on Crown Prosecution Service to improve rape statistics after NationalWorld complaint

The UK statistics watchdog has asked the Crown Prosecution Service (CPS) to make improvements to its rape statistics after NationalWorld raised concerns they inflate the rape charge rate to make the prosecution record appear better than it is. 

Ed Humpherson, director general for regulation at the Office for Statistics Regulation (OSR), a non-ministerial government department which provides independent regulation of all official statistics in the UK, wrote to the CPS earlier this month upholding concerns raised by NationalWorld about the transparency and quality of the figures it releases. 

His letter followed a complaint by NationalWorld outlining our concerns that the statistics were misleading. The CPS has subsequently made changes to the way its data is presented.

Currently, the CPS only produces unofficial statistics – although they are still widely cited by MPs and the media without the non-official status being made clear – so is not bound by the Code of Practice for Statistics, which the OSR enforces. But Mr Humpherson urged it to voluntarily sign up to the code, while stating their statistics in their existing form did not “meet our expectations for intelligent transparency”. 

Every three months the CPS releases figures that it previously said showed the volume of rape cases referred to it by police; the proportion of those cases in which it authorises charges to be brought against a suspect; the volume of prosecutions pursued; and the proportion of prosecutions that result in convictions. 

These statistics previously appeared under a heading that simply read ‘rape’. However, the figures do not in fact count actual charges, prosecutions or convictions for rape offences. Instead, the CPS uses a ‘rape flag’ system. Every time police forces refer a case to it for a charging decision, prosecutors apply a rape ‘flag’ against the case in its administrative system, so the outcomes can be monitored. But this flag remains in place no matter what charges are actually brought. 

A single case against a single suspect referred to the CPS may concern multiple different offences – for instance, one alleged offence of harassment and one of rape. If the CPS pursues a charge for harassment but not rape, this will appear as a rape-flagged charge. And if they pursue both charges but only secure a conviction for harassment, it is logged as a successful rape conviction. 

NationalWorld has spent several months investigating how many of the rape and sexual assault suspects official Home Office data claims have been charged were actually charged with offences other than rape or sexual assault. In many cases, these offences have been non-sexual, including common assault and harassment. While it is data recorded by the police that has been the focus of our investigation, the CPS’s system means such cases will likely be captured as rape prosecutions and convictions, despite many defendents never facing a rape charge. 

Official figures on prosecutions and convictions in the courts are maintained by the Ministry of Justice (MoJ) – and analysis of these shows the CPS has a much worse record than its own internal statistics suggest.

CPS figures covering October to December 2022 show there were 712 rape-flagged prosecutions completed, alongside 440 convictions, giving a conviction rate of 61.8%. But according to the MoJ, there were 1,935 rape charges laid before the courts in the same period, and 624 convictions, giving a conviction rate of 32.2%. When measuring based on the number of defendants convicted where rape was the most serious offence (some defendants face multiple charges), the conviction rate drops to 29.9%. 

Previously, the CPS’s rape flag system was outlined at the very bottom of the webpage on which its statistics were published. In an ‘About CPS data’ section, readers were also told that the figures are not official, and that the MoJ and Home Office produce official data instead – although the website did not contain a link to these sources. 

In response to our complaint about the rape flagging system, Mr Humpherson told the CPS the “methodological limitation is crucial for interpreting the end data and must be made clearer to users, through a prominent caveat alongside the data”. He also recommended the CPS include clearer and more prominent definitions next to the rape flagged statistics. 

“It is never our aim to inhibit transparency and we welcome the fact that CPS is putting these data into the public domain to inform public debate,” Mr Humpherson wrote. “However, the distinction between official statistics and other data may seem artificial to many users of these data. Releasing data without sufficient context and clarity may limit the extent to which they can usefully inform public debate and does not meet our expectations for intelligent transparency.

“We therefore strongly recommend that CPS takes steps to voluntarily apply the Code of Practice for Statistics. This will demonstrate to users that CPS seek to commit itself to the highest principles and practices of trustworthiness, quality and value embedded in the Code.”

In a letter of response, Max Hill KC, Director of Public Prosecutions at the CPS, accepted all of the recommended changes, and said these had been reflected in the latest round of figures, which were released on 20 July.

The data table concerning rape is now labelled “rape flagged cases”, while a note states: “The following tables show flagged cases and their outcomes. Due to the nature of CPS systems this will include cases where the eventual outcome may not relate to the flagged offence.”

Mr Hill wrote: “Whilst we have been able to address these suggested changes at pace, we know that there is further work that the CPS can do to continue to strive for intelligent transparency.

“We will be looking carefully at a future voluntary application to the Code of Practice as we continue to seek to commit ourselves to the highest principles and practices of trustworthiness, quality and value that are embedded in the Code.”

In a statement to NationalWorld, a CPS spokesperson said: "We welcome this advice and the suggested changes as we aim to maintain our commitment to transparency on prosecution performance. The way we use our data has allowed us to track and monitor trends over many years and is presented with caveats about its limitations." 

The spokesperson also said that it had looked at its statistics in detail, and found 80% of rape flagged cases go on to be charged as rape, while 16% are another serious sexual offence. However NationalWorld has previously submitted a Freedom of Information request to the CPS asking how many actual rape charges it takes forward, and been told it was not possible to extract this data. It is unclear why it has now been able to do so.