‘Fiddling the figures’, ‘Cooking the books’ or ‘Good housekeeping'?

Readers might enjoy this submission by Dr Rodger Patrick to the Justice Committee.

Dr Patrick is recognised as expert in the field of exposing "Gaming" or the manipulation of crime figures by police forces to reflect improved performance.

www.publications.parliament.uk/pa/cm201012/cmselect/cmjust/writev/foi/foi.pdf

It is also worth having a look at the West Midlands Police submission on behalf of ACPO later in the link above.

Ironic that Dr Patrick used the FOI to expose the failings of the IPCC by accessing internal reports from the West Mids Police and the Chief Constable now suggests that submissions raking over investigations conducted by the IPCC should be barred.

Written evidence from Dr. Rodger Patrick

Executive Summary:

The provisions of the Freedom of Information Act 2000 were utilised to access the evidence to
substantiate the Doctoral Study, and subsequent academic articles, on the impact of Performance Management on the Police Service in England and Wales. Reflections on the experience of using the legislation as an academic device are outlined in the submission, as are examples of the type of information disclosed. It was established that the Act was a powerful means of accessing information with the potential to facilitate greater accountability. However, on its own it cannot ensure reform and further legislation may be necessary to realise its overarching objectives.

1. Rationale for utilising the Freedom of Information Act (FOI) as a research tool:

1.1 Participant observations as a Chief Inspector in the West Midlands Police had identified ‘gaming’ behaviour as an unintended consequence of Performance Management introduced as a means of improving Police governance and accountability. However earlier disclosures of similar perverse behaviours by practitioners (Young 1991) had been unable to establish an empirical pattern and was weakened by a lack of corroboration. The Police Federation had also responded to their members’ concerns by highlighting ‘gaming’ behaviours at their national conference in 2007 but were accused by the then Policing Minister Tony McNulty of ‘over‐egging the cake’.

The introduction of the Freedom of Information Act provided a means of accessing unpublished documents and detailed official statistics, thus addressing these methodological limitations.

1.2 Observations had identified four distinct categories of perverse behaviours designed to improve performance by unethical means, or give the false impression performance had improved. Such behaviours are referred to colloquially as ‘fiddling the figures’, ‘cooking the books’ or ‘good housekeeping’. In the academic literature they are referred to as ‘gaming’ behaviour and there is a body of theory on the subject (De Bruijn 2001 2007, Le Grand 2003 and Bevan & Hood 2006)

1.3 The four categories of ‘gaming’ behaviours identified are defined as follows:

• Cuffing: The under‐recording of reported crimes, the term being derived from the magician’s
art of making objects disappear up the sleeve or cuff (Young 1991).

• Nodding: This involves collusion between officers and suspects to admit to large numbers of
offences, usually whilst in prison after sentence, in return for favours such as reduced
sentences, access to partners, drugs or alcohol. The term is used to describe the act of a
prisoner pointing out or ‘nodding’ at locations where they claim to have committed
offences. (Wilson et al 2001:63)

• Skewing: This involves moving resources from areas of activity which are not subject to
performance measures in order to improve performance in areas that are monitored for
control purposes (Rogerson 1995).

• Stitching: This includes a variety of malpractices designed to enhance the strength of the
evidence against a suspect in order to ensure the desired criminal justice outcome.
Fabricating evidence or stitching someone up are forms of this behaviour.

1.4 All of these ‘gaming’ behaviours were operating in the West Midlands Police during the period 1996 to 1999 and had a major impact on the force’s performance (Patrick 2004, 2009 2011 2012a 2012b). In combination they can be regarded as a ‘perverse policing model’. The statistical profile of the force during this period provided a template for identifying similar practices in other forces.

1.5 Various ‘events’, some in response to the behaviours being uncovered by regulators or specific incidents which had the potential to become scandals, had resulted in the force conducting reviews of the concerning behaviours. The resulting documentation, much of which was completed prior to the introduction of the Freedom of Information Act (FOI), was never intended for public consumption. However when presented with a FOI request, the force released the documents. Many of these reports quantified the scale of the behaviours and by supplementing this information with analysis of national performance data it was possible to calculate the extent to which ‘gaming’ behaviour distorted results.

2. Use of the FOI Act to conduct an academic survey:

2.1 The introduction of a False Reporting Policy in the West Midlands Police in 2004 corresponded with a marked reduction in reported crime. It was argued that this represented a return to an evidential crime recording standard and the associated ‘cuffing’ (Patrick 2011). This demonstrated a flaw in the National Crime Recording Standard introduced in April 2002 with the stated objective of standardising crime recording procedures and eliminating such distortions. In order to establish if other forces had experienced similar reductions, when they introduced such a policy, all forces in England and Wales were sent a questionnaire under the provisions of the Act asking them to supply copies of any false reporting policy.

2.2 They were also asked to provide any analytical assessment of the problem of false reporting
completed under the National Intelligence Model (NIM) and any Impact Assessment completed on the policy under the provisions of the Race Relations (Amendment) Act. It was hoped the National Intelligence Model assessment would quantify the scale of the potential threat from what was believed to be the bogus reporting of crime in pursuit of fraudulent insurance claims.

The Impact Assessments should have been completed as the re‐introduction of greater discretion into the crime recording process could be influenced by officer bias. Again some quantifiable results were being sought.

2.3 A 100% response rate was achieved. The identification of those forces operating ‘false reporting’ policies enabled a statistical analysis to be completed and this supported the hypothesis that the crime recording standard had reverted back to its pre NCRS interpretation. Unpublished data on the number of crimes which had been declassified as crimes i.e. ‘no crimed’, was released by the Home Office and this strengthened the academic argument.

2.4 Only two of the forces to have introduced false reporting policies had conducted NIM
assessments on the threat and neither supported the introduction of the policy. None of the
identified forces had carried out comprehensive impact assessments on the re‐introduction of
discretion into the crime recording procedures although one had identified the potentially
discriminatory consequences of the refusal to issue crime numbers to those reporting passports
stolen.

2.5 The ‘evidence of absence’ is a valuable facet of the Act and proved valuable when exploring why the Home Office ceased publishing the validation tests they carried out on the police recording rates in the British Crime Survey (BCS). A FOI request established that they ceased asking the relevant questions in the 2006/7 survey despite the fact that the results in 2005/6 indicated ‘under recording’ was increasing (Walker, Kershaw and Nicholas: 2006:52).

3 Accessing Statistics:

3.1 In order to quantify the extent of ‘nodding’, associated with abuse of the Offences Taken into
Consideration (TIC) procedures, it was necessary to analyse the impact of preventative action on the number of offences detected as TICs. The Home Office again released the data in response to a Freedom of Information request and the scale of the abuse could be estimated from the fall in TICs after the incident or scandal which stimulated the remedial action. A consistent and compelling pattern was evident in all the cases examined (Patrick 2012a).

4 Accessing Confidential Documents:

4.1 The information contained within sensitive and confidential documents had a major impact on the direction of the research and the conclusions reached. Whilst participant observation and
knowledge of procedures was a great benefit, the willingness of organisations to comply with the
statutory requirements was commendable. The release of a confidential force inspection report
completed by Her Majesty’s Inspector of Constabulary (HMIC 1998) stimulated interest in the way regulators addressed ‘gaming’ when it was uncovered. It was clear from this document that HMIC were aware as early as 1998 that post sentence admissions, Prison Write Offs, were being converted to TICs and that the practice was prone to abuse.

4.2 A draft report on Nottingham Constabulary, following the Chief Constable’s admission that the force was struggling to cope with a marked rise in gang related murders, revealed that the redeployment of officers from specialist HQ squads to local policing units had not gone well. This
‘skewing’ of resources in favour of more affluent areas had been identified in the West Midlands and it appeared, in both cases, that it could have contributed to the rise in gang related crime or
curtailed these forces’ ability to respond to the emerging threat. The paragraph referring to this was omitted from the published version of the report.

4.3 Confidential documents released by one force facilitated an assessment of the capabilities of the Independent Police Complaints Commission. The documents challenged their findings on an
important investigation triggered by the death of a child (Patrick 2012b).

5. The Appeals Process:

5.1 In order to pursue the hypothesis that regulators may be pressurising forces to introduce
‘gaming’ type behaviours to improve their performance, information was requested from the Police Standards Unit (PSU). Specifically sought were their assessments and recommendations on West Yorkshire and Nottinghamshire police forces. It was known that both forces had experienced ‘gaming’ type problems during the late 1990s/early 2000s and the remedial action no doubt contributed to their low station on the national league tables. Their performance improved markedly after Police Standard Unit intervention and examination of their performance profiles suggested ‘nodding’ and ‘cuffing’ had been resurrected. The request for full disclosure was refused and this led to an appeal to the Information Commissioner.

5.2 Whilst the Information Commissioner found in favour of the appeal, the process took over three years to complete. This is believed to be due to a lack of resources at the Information
Commissioner’s Office and is something the Committee may wish to comment on.

6. The scope of Organisations subject to the provisions of the Act:

6.1 The Act was again to prove invaluable when studying the behaviour of the regulators. In 2006 problems with the impact of ‘gaming’ on the reliability of detection data appeared to be behind the decision to audit a range of detections which did not involve the charging of a suspect. Contrary to long established practice the results were not published by Her Majesty’s Inspector of Constabulary (HMIC). However, despite the fact that HMIC are not subject to the provisions of the Act, they responded to a request from an unknown applicant and published the results on their web site. The exclusion of HMIC from the Act may be something the Committee may wish to review.

6.2 The audit of detections highlighted major quality failings, including insufficiency of evidence, in relation to cautions, penalty notices and informal warnings. Such deficiencies would have
implications for the Criminal Record Bureau and the Association of Chief Police Officers (ACPO) who administer the Police National Computer (PNC). The use of the FOI to follow the documentary trail was not an option in relation to ACPO as this organisation is not covered by the Act. Again the Committee may wish to review the organisations covered by the legislation.

7. The Operating Philosophy of the Information Commissioner:

7.1 The evidential ‘blind spot’ in relation to ACPO was partially overcome when they wrote to the Information Commissioner admitting that the audit of detections had uncovered potential breaches of the Data Protection Act. All the documentation in relation to this exchange was released by the Commissioner’s Office. The subject of the correspondence was encapsulated by the following disclosure:

“From my perspective I am most concerned that individuals were not being informed that
they were considered to be the perpetrator of an offence even though this did not involve a
legal process, especially if such information could be used in future Enhanced Disclosure
relating to them. This clearly breaches the requirement of the first data protection principle
that the processing of personal data must be done fairly. I am also worried by the sufficiency
of evidence used. If a police force is going to label an individual as the de facto perpetrator
then they must have a good objective reason for doing so. Not having this could lead to a
record being viewed as inadequate or inaccurate (breaches of the third and fourth principles
respectively).” (Information Commissioner 26.3.2007: Unpublished)

7.2 Whilst ACPO were less than candid about the scale and nature of the ‘insufficiency of evidence’ uncovered by the audit, the decision by the Information Commissioner not to alert the public to the risk they had been exposed to demonstrated a tendency towards a ‘professional body’ approach to regulation, as opposed to a ‘democratic model’ where information is made available to the public in order for them to hold officials to account:

“I would prefer to work with chief officers to ensure compliance. I would like to know more
detail about how this has come about and what action is being taken to ensure future
compliance” (Information Commissioner 26.3.2007: Unpublished)

7.3 The reply from ACPO disclosed the fact that the Home Office and HMIC were also aware of the nature of the problem:

“Please rest assured that the issues that have come to light as a result of the recent Association of Chief Police Officers’ and HMIC audits have been taken extremely seriously by the police service. To this end a series of meetings have taken place in fast time with all relevant parties, including the Home Office, to consider how best to address the concerns that you raised to which we are alive” (ACPO letter to the Information Commissioner 10.4.2007: Unpublished)

8. Conclusion/Discussion

8.1. The provisions of the Freedom of Information Act were an invaluable research tool overcoming a number of methodological limitations faced by previous studies. Whilst the findings could be damaging to the State, thus defeating the desired objective of enhancing public trust, research in this area suggests that such loss of trust can be transient (Van de Walle et al 2008). So in the medium to long term, greater transparency will lead to reform and better governance thus vindicating those who introduced the legislation.

8.2 It should also be borne in mind that the study referred to benefited from an element of surprise and many of the documents relied upon preceded the enactment of the legislation. Anecdotal evidence would suggest such candid documents are being shredded. It could also be surmised that they may not be commissioned at all. If this were to be the case then it could be argued the Act has been counterproductive.

8.3 The Information Commissioner appeared to give voice to this concern in Plymouth City Council Vs BBC (2006) (ICO Ref FS50082254). In this case the BBC had requested full disclosure of the documents submitted to a serious case review into the death of a child by individual agencies. The Information Commissioner judged that the disclosure of the reviews conducted by individual agencies may deter them from being open thus limiting the ability to learn from experience. This rational was challenged by the deliberations following the death of Baby P.

8.4 Whilst the reluctance of Chief Officers to disclose relevant information to regulators or other
forms of official inquiry can be addressed by legislation and conferring greater powers on
investigators, this will be to no avail if the regulators themselves are not committed to making their findings public. The failure of the Financial Services Authority was a damning indictment of this approach. Structurally this flaw may be alleviated by aligning the regulators to Parliament and its Scrutiny Committees may be the most appropriate bodies to oversee their work.

8.5 Another consideration raised by the study was the means by which information uncovered by Freedom of Information requests is disseminated to the general public. Many of the documents disclosed were not made public or easily accessible on the web sites of the originating organisations, including the Information Commissioner’s Office. This is seen as good practice but is not a statutory requirement. The media is one outlet but they have their own agenda and move on quickly to the next breaking story (Howard & Waisbord 2004). Academic articles are one outlet but publication is slow. The official regulators appear to be the most obvious recipient but as can be seen from the evidence presented appropriate action may not follow.

8.6 In relation to the information uncovered by this study the most appropriate outlet would appear to be the Parliamentary Scrutiny Committees. Certainly something needs to be done for those citizens who have police records which are not justified. Witness testimonies and analysis of the Home Office data would suggest the same problems are occurring with offences dealt with as Restorative or Community Resolutions. It may be worth the Scrutiny Committees considering making themselves more accessible to un‐solicited submissions.
January 2012

References:
Bevan & Hood (2006). What’s Measured is What Matters: Targets and Gaming in the English Public Health Care System. Public Administration Vol. 84, No 3, 2006 (517‐538)
De Bruijn, H. (2002). Performance measurement in the public sector. (London Routledge
Her Majesty’s Inspector of Constabulary (1998). An Audit Report by Her Majesty’s Inspector of
Constabulary completed on the West Midlands Police in December 1998.Not published: available via Freedom of Information request to the West Midlands Police.
Howard T. & S. R. Waisbord (2004). Political Scandals and Media Across Democracies, Volume II
AMERICAN BEHAVIORAL SCIENTIST, Vol. 47 No. 9, May 2004 1143‐1152
Le Grand J. (2003). Motivation, Agency and Public Policy. Oxford University Press.
Patrick, R. (2004). The Lot of the Poor is Unlikely to Improve Until They Have a Greater Say in Their ‘lot’: Going Local and the Impact on the Distribution of Public Services. Vista Perspectives on Probation and Criminal Justice & Civil Renewal Vol. 9 No. 3 2004.
Patrick R. (2009) Performance Management, Gaming and Police Practice: A Study of Changing Police Behaviour in England and Wales During the Era of New Public Management. University of
Birmingham. PhD Thesis.
Patrick, R. (2011). ‘Reading Tea Leaves’ an assessment of the reliability of police recorded crime
statistics. The Police Journal, Volume 84 (2011) pgs. 47‐67.
Patrick, R. (2012a). ‘A Nod and a Wink’: Do ‘gaming practices’ provide an insight into the
organisational nature of police corruption? (The Police Journal. Accepted for publication,11.3.11
Ref.: Ms. No. PJ‐D‐10529R1).
Patrick, R. (2012b). 'Public Champions or Protectors of Professional Interests?' Observations on the performance of those bodies entrusted with the regulation of the Police Service during the era of New Public Management’ (The Police Journal. Accepted for publication 17.6.2011 Ref.: Ms. No. PJ‐D‐11536R1)
Rogerson, P. (1995) Performance Measurement and Policing: Police Service or Law Enforcement
Agency? Public Money & Management, October‐December 1995
Van de Walle Steven., Steven Van Roosbroek and Geert Bouckaert (2008) Trust in the
public sector: is there any evidence for a long-term decline? International Review of
Administrative Sciences 2008; 74; 47
Walker A., Kershaw C. & Nicholas S. (2006). Crime in England and Wales 2005/2006 Home Office, London
Wilson, D., J.Ashton, & D. Sharp (2001). What Everyone in Britain Should Know About the Police. Blackstone.
Young, M. (1991). An Inside Job. Policing and Police Culture in Britain. Oxford: Claredon Press