Mark Pont to Daniel Shaw: Operation of police powers under the Terrorism Act 2000

Dear Daniel

Operation of police powers under the Terrorism Act 2000

As you are aware, we recently reviewed Home Office (HO) statistics on the Operation of police powers under the Terrorism Act 2000 against the Code of Practice for Statistics. Following a constructive conversation with the team responsible for producing these statistics, I am pleased to confirm that they should continue to be designated as National Statistics.

We found many positives in our review, particularly in how your team presents these statistics, which enhance their value and quality. These include:

  • The use of infographics and flow charts to illustrate the criminal justice process and the outcomes for those arrested. These graphics are simple but highly effective at summarising the data and conveying visually an overall story that could be hard to bring together across a written publication.
  • The team is taking steps to ensure its statistical bulletins meet the new accessibility requirements for public sector organisations, such as including text to describe charts, which can be read by those using screen readers, and making changes to the standard HO colour palate that is used across many statistical publications. This work will ensure a broad range of users is able to access the statistics easily. We also welcome that there is an ongoing discussion in HO regarding publishing statistical outputs in HTML format to aid accessibility.
  • The user guide that accompanies this statistical publication presents insightful, clear information about the different data that make up the statistics and how the quality of these is assured. We particularly like the presentation of this information in tables (pages 2-4 of the current user guide), as we feel it makes it very easy for users of the statistics and guide to find this information.
  • The user guide also provides helpful information about how data are collected and resulting limitations on the statistics. For example, it is clear that ethnicity and nationality data may be based on the officer’s best judgement, rather than the actual ethnicity or nationality of the person being arrested. The team is aware that some users of the statistics would like ethnicity data to be broken down into more-detailed ethnic groups and is exploring with data providers whether this is possible.
  • In response to user feedback, HO has improved the timeliness of these statistics. Previously data were published 6 months after the period to which they referred; HO has worked with data providers to reduce this to 3 months.


Our review also identified several ways we consider you could further enhance the trustworthiness and value of these statistics:

  • The pre-release access list for these statistics is very long, and it is not clear to us that it is essential for all of the named roles on the list to receive the statistics 24 hours in advance of their general release. The trustworthiness of government statistics is in part determined by them being known to be produced independently from outside influence. This perception can be damaged when pre-release access is too wide. We were pleased to hear that your team plans to review the list of individuals who receive pre-release access to these statistics, with a view to reducing the total number before the next publication.
  • The statistical bulletin and the user guide make clear that the data in these statistics are regularly revised, and that each release contains data that are correct at the time of publication. While the team told us that the sizes of revisions between publications are small, it would be beneficial for the team to do some quantitative analysis that illustrates the size of typical revisions between publications of the statistics. This would better inform users about the size of revisions and help them understand how data might change in the future, which may impact how they analyse or interpret analysis of the statistics.
  • Although there is substantial evidence of the team making improvements to the timeliness and content of these statistics in response to user needs, the user base with which the team engages remains narrow. We think that the team’s suggestion of contacting academics working in this space as a way into understanding the wider user base is a good idea.

Thank you to your team for their positive engagement during this review: we look forward to continuing to engage with you and the team on these and other statistics. Please do not hesitate to get in touch if you would like to discuss any aspects of this letter further.

I am copying this letter to John Flatley, Programme Director for Crime and Policing Statistics at HO, and Jodie Hargreaves, Crime and Policing Analysis, HO Analysis and Insight.

Yours sincerely


Mark Pont

Assessment Programme Lead


Related Links

Statistics on the Operation of Police Powers under the Terrorism Act 2000 and other legislation


Response from Daniel Shaw to Ed Humpherson: Parliamentary question response

Dear Ed,

RE: Parliamentary question response on figures regarding those subject to No Recourse to Public Funds policy.

 Thank you for your letter of 22 June 2020 where you raised an issue with regards to how the Home Office had responded to a parliamentary question requesting data on the numbers of people were given leave to remain in the UK subject to the No Recourse to Public Funds (NRPF) condition.

I fully acknowledge that the wording used in this response was unclear and did not correctly reference the Code of Practice for statistics, nor was it sufficiently transparent with regards to the data held by the department.

In responses to parliamentary questions relating to the Code of Practice for statistics standard Home Office practice would be to clear such responses with a statistician. In this case the response was issued without this taking place. I have taken this opportunity to raise this with operational and policy colleagues, reminding them that this should be a standard part of the clearance process.

With regards to the specific question how many people were given leave to remain in the UK subject to the no recourse to public funds condition in 2019. Home Office administrative data only captures information on whether visas are subject to NRPF conditions for in-country extensions. The information is not available for applications from overseas which form the majority of applications. Therefore, no complete figures of visas subject to NRPF conditions can be produced.

The majority of migrants visiting, studying, working or joining family in the UK are subject to a NRPF condition until they have obtained indefinite leave to remain. Migrants who are here without leave are also subject to NRPF.

I recognise that other organisations have tried to estimate the total population within the UK to which NRPF restrictions will apply. Estimates of this population will likely have been based upon our published statistics However, such estimates will only provide the numbers of people whose issued entry clearance visa or granted leave to enter/remain include the NRPF clause, which as noted will apply to most of the foreign nationals who come to the UK.

I do not feel that it is of practical application for the Home Office to produce an estimate of the total population subject to NRPF present in the UK at any one time. This is for a number of reasons:

  • The information captured by the Home Office doesn’t take into account individuals who have left the country or all those in the UK without lawful status. Therefore, they cannot be used to accurately measure the resident population at any one time.
  • The NRPF condition applies to millions of visa applications, the vast majority of which are visitors and other temporary migrants who would have no requirement to access funds during their stay.
  • The size of the total resident population of individuals subject to NRPF at any one time will therefore simply be driven by the total numbers of visas issued. It would not provide any information on whether the conditions were in any way detrimental to those individuals.

There is a clear public interest in NRPF with regards to the numbers of individuals where the policy has prevented access to public funds if this subsequently were to give rise to hardship.

Exceptions are already in place for some group of migrants, such as refugees, protected persons and those granted discretionary leave, or families here on the basis of family life/Article 8 where the family would otherwise be destitute. Individuals on family life and human right routes can also apply to have their no recourse to public funds condition lifted if their financial circumstances change.

Only the benefits listed in immigration legislation are classed as public funds for immigration purposes. Some benefits may still be accessed where they are based upon national insurance contributions such as contributory based employment and support allowance and statutory sick pay.

Local authorities may also provide basic safety net support, regardless of immigration status, if it is established that there is a genuine care need that does not arise solely from destitution. Examples include where there are community care needs, migrants with serious health problems or family cases where the wellbeing of a child is in question.

Administrative data held by the Home Office will only relate to the status of immigration decisions and will not provide information on individual interactions with the benefits system or with other organisations, such as Local Authorities. Nonetheless, I am committed to further investigate the administrative data we hold to assess whether it can provide any meaningful information on the issue of hardship specifically and will update your team as soon as practically possible.

Yours Sincerely,

Daniel Shaw


Related Links

Sir David Norgrove response to Stephen Timms MP

Stephen Timms MP to Sir David Norgrove

Ed Humpherson to Daniel Shaw: Parliamentary question response

Ed Humpherson to Daniel Shaw: Parliamentary question response

Dear Daniel

I am writing regarding issues that have come to my attention linked to a parliamentary question response asking for information on how many people were given leave to remain in the UK subject to the no recourse to public funds condition in 2019. I thank you for engaging with my team on these matters.

I am pleased to hear that you are taking steps to improve the processes within the Home Office when responding to parliamentary questions which relate to statistical matters. I will expect to see clearer explanations and clarity over the standards required when producing data and statistics, as set out in the Code of Practice for Statistics, in the future. Whether through parliamentary question responses or any other communications from the Home Office regarding the ability to meet user need for data and statistics, fuller explanations should be provided, being transparent about the reasons for the decisions made and any constraints.

As I understand, you do not consider that an accurate answer of sufficient quality can be provided at this time. I acknowledge your professional judgement on this matter. I believe you are taking action to further investigate these data quality issues. I welcome that development and ask you to work with urgency to take this forward.

The data requested relates to an important policy area and there is clear demand for information on this topic. While you progress work to understand data quality and consider what analyses may be possible – taking account of user needs – I suggest you look to provide alternative data or estimates that may help inform the issue. Recognising that these estimates may have limitations and be of poorer quality, users should be provided with clear guidance to support appropriate use of this data.

We will continue to engage with you and your team as work progresses.

Yours sincerely

Ed Humpherson

Director General for Regulation


Related Links

Sir David Norgrove response to Stephen Timms MP

Stephen Timms MP to Sir David Norgrove

Home Office statistics on Police Workforce and Police Powers and Procedures

Dear John


As you know, we have recently conducted our review of the compliance of two Home Office statistics against the Code of Practice for Statistics. These statistics are:

These statistics are especially important given the continued focus on policing in public debate. Thank you for the positive engagement from you and your team throughout our review process. I am pleased to confirm that these statistics should continue to be designated as National Statistics.

We found many positive aspects in the way that the Home Office produces and presents these statistics, including:

  • Neutral but insightful commentary is provided alongside the statistics throughout each of the sets of published bulletins.
  • Presentation of the statistics in both sets of publications has been improved as a result of user feedback. Data tables are now more user-friendly and allow for easier and quicker analysis.
  • The supporting user guides for both sets of statistics explain the data limitations in a transparent way, which increases the trustworthiness and value of the statistics and helps users make better-informed decisions.
  • The statistics team actively engages with police forces to better understand how the data used in the production of the statistics are obtained and recorded. These discussions mean the statisticians have insight into the constraints and difficulties that some police forces face in providing high quality data. The team has developed data collection templates with guidance and inbuilt quality checks that aim to reduce the burden on the forces and to ensure a higher level of data quality.
  • The team shows an ongoing commitment to improving the statistics and ensuring that they help to answer relevant policy questions: it has introduced new statistics, clearly labelled as ‘experimental’, which means data are available, but those using the statistics can be clear that they are still in development.
  • Finally, the team continues to consider new ways of presenting data to add value for those using them.

We identified two areas where we consider that the team should enhance the value of the statistics:

  • We are aware that there is demand for geographic breakdowns of these statistics below England and Wales level: for Wales as a whole, for example. To help people and organisations who need to understand the policing landscape across specific nations or regions, we expect the team to understand this demand and to meet it as best it can. This could include clearer signposting in the statistical bulletin to supplementary data tables that contain breakdowns of the overall figures; making sure that all supplementary data tables include separate totals for England and Wales; and drawing out any differing trends observed between nations within the statistical commentary.
  • We suggest that the team include information in the police workforce statistical bulletins and/or user guide that provides more clarity on the scope of the statistics, detailing the types of officers not currently included in the published numbers; those working in the National Crime Agency (NCA), for example. Providing this information will help users of the statistics understand exactly what the numbers presented to them represent.

We are aware that the team is currently conducting a user consultation to seek feedback on a proposal to extend the coverage of the police workforce statistics to cover NCA officers and that the team is exploring with police forces the potential for presenting new statistics on the number of people ‘released under investigation’, after policy changes in 2017. We welcome this focus on user-led continuous improvement.

Alongside these initiatives, we were very pleased to hear from you on 28 January that the statistics team is working to develop a new data stream to allow Home Office to report quarterly on progress against the UK Government’s intention to recruit an additional 20,000 police officers. We commend the collaborative approach of your team, which is working with National Police Chiefs Council and individual police forces to provide these statistics, and your information note for users, which clearly communicates the proposed methodology for calculating the initial workforce baseline and how progress will be reported on. We welcome the development as a really good example of seeking to produce statistics that address topical issues and contribute to improving public debate.

We look forward to continuing to engage with you and the team on these and other statistics.

I am copying this letter to Jodie Hargreaves (Crime and Policing Analysis, Home Office Analysis and Insight).

Yours sincerely

Mark Pont
Assessment Programme Lead


Related Links:

Home Office letter to Ed Humpherson (January 2020)

Reporting the costs of domestic abuse

Dear Mike,

Home Office report: the economic and social cost of domestic abuse in England & Wales

Robert Cuffe, Head of Statistics at BBC News, recently contacted us about the estimated cost of domestic abuse that was published in your report: The economic and social cost of domestic abuse and the associated press release that announced the publication of the draft Domestic Abuse Bill.

Mr Cuffe was concerned that the analysis in the report has the potential to triple-count the cost of abuse by using an unusual definition of the costs associated with abuse in a specific year. He also suggested that the way the results were communicated does not make the definition clear to the lay reader.

We welcome your publication of this new analysis and the transparency of the methods, but share some of the concerns raised by Mr Cuffe about how the information has been presented, particularly in the associated press release.

Your report on the economic and social cost of domestic abuse aims to estimate the costs of domestic abuse in England and Wales for the year ending 31 March 2017. It estimates a total cost of £66.2 billion, made up of three elements:

  • £47.3 billion: the estimated cost to those experiencing domestic abuse for the reduction in their quality of life due to the physical and emotional harms they experience
  • £14.1 billion: the estimated cost to the economy of lost output
  • £4.7 billion: the estimated cost of public and other services such as private civil and criminal legal representation and charitable services

The estimated costs for reduced quality of life and lost output represent the notional loss to victims and the economy of domestic abuse over the time they experienced the abuse. Your analysts confirmed they calculated those costs over a three-year period, which was taken as the average time victims experience abuse. The first two estimates therefore represent the three-year total costs associated with victims of abuse who were identified during the year ending 31 March 2017, not the total in-year costs. The costs associated with public and other services include a mixture of estimates calculated over a three-year and a one-year period.

Your analysts assured us that they were aware of the risk of including the same costs more than once using this method (if the estimates were repeated each year) and that they have no plans to publish these estimates on annual basis.

The report was written for a technical audience with some understanding of the methods. The press release had wider circulation and was written for a lay reader. It stated that the report ‘reveals [domestic abuse] cost England and Wales £66 billion in 2016 to 2017’. The press release did not make clear that the costs were estimated; that the costs were not all in-year costs; or that most of the costs were notional, rather than incurred costs.

Communications staff have a difficult job to explain technical material like this concisely. It is vital that analysts responsible for the work are able to advise on press releases to ensure that statistics are reported accurately and are not misrepresented. We understand that, in this case, some of the analyst’s contributions that would have aided interpretation were not incorporated in the final version of the press release.

I am copying this letter to Andy Tighe, Director of Communications, Amanda White, Deputy Director for Crime & Policing Analysis, and David Blunt, Head of Profession for Statistics, all at the Home Office.

Your sincerely

Ed Humpherson
Director General for Regulation