Public Administration Committee
Oral evidence: Crime Statistics, HC 760
Wednesday 11 December 2013
Ordered by the House of Commons to be published on Wednesday 11 December 2013
Written evidence from witnesses:
Members present: Mr Bernard Jenkin (Chair); Alan Cairns, Paul Flynn, Robert Halfon, Kelvin Hopkins, Priti Patel, Mr Andrew Turner
Questions 136-283
Witnesses: Professor Stephen Shute, Chair and professor Mike Hough, Member, Crime Statistics Advisory Committee, gave evidence.
Q136 Chair: Good morning, everyone. To our witnesses, may I welcome you to this session on crime statistics? We have a number of questions to get through; we will try to ask short questions, and if you can give us short answers it would help us enormously. If you are going on a bit too long, I hope you will not mind if I pull you up or try to get you to focus on the question that is being asked. Could I first of all ask you to identify yourselves for the record, please?
Professor Shute: My name is Stephen Shute. I am Professor of Criminal Law and Criminal Justice at the University of Sussex, where I also head up the School of Law, Politics and Sociology. I am Chair of the Crime Statistics Advisory Committee and appear in that capacity.
Professor Hough: My name is Mike Hough. I am Professor of Criminal Policy at Birkbeck and a member of CSAC, and I have had long association with the British Crime Survey, crime counting and police research.
Q137 Chair: You will be aware of our previous evidence session. I particularly want to start by asking what academic and independent analysis CSAC can show about the extent of misrecording of crime data.
Professor Shute: Perhaps I could start by addressing that question. We rely on the work of others, in the sense that we do not commission academic work directly. We have to rely on the body of academic work that has been done by others, although we work closely with Her Majesty’s Inspectorate of Constabulary, and, indeed, the Chief Inspector of Constabulary is a member of the committee. One of our functions is to advise HMIC on how it does its work in relation to crime data integrity. We receive reports from them on the inspections they have done and plan to carry out in this area, and offer input into that inspection process. Indeed, one of our members has been nominated to sit on the external reference group for the latest planned inspection of crime data integrity that will be run by HMIC next year.
Q138 Chair: What is your opinion of the extent of the misrecording of police recorded crime?
Professor Shute: It is hard to measure the extent. Misrecording of crime can come about for a number of reasons.
Q139 Chair: Can you just answer the question? What is your view about the extent? However hard it is to measure, what judgment do you arrive at, on the basis of, I accept, relatively limited information?
Professor Shute: We have no really hard evidence. We have seen a divergence between the two measures of crime: police reported crime and the Crime Survey for England and Wales. That might allow one to draw some assumptions about what the level of misrecording of crime might be. We have the inspections that HMIC has done of Kent and we have the inspections that the Audit Commission did some years ago about misrecording. One could assume that one will never be able to reach the point where you get 100% of crime recorded, but there are indications from HMIC that using the methodology that they use—which does not capture all crime incidents that come to a police force; it captures those that come into control rooms—in many forces we are getting 80% or 90% of crimes that are reported to police within the Home Office counting rules being recorded appropriately by the police.
Professor Hough: I think we know quite a lot about the extent of under-recording, which is slightly different from misrecording. If you take a view on long-term trends, in the 1980s and 1990s, around about 50% or 60% of reported crimes found their way into police statistics. Then with the introduction of the changes at the end of the 1990s and early 2000s, that was pushed up to 90%—90% of reported crimes got into police statistics. From 2007 or 2008 onwards, that proportion has been falling, so now about 70% of reported crimes end up in police statistics. Clearly the shortfall is made up partly of misrecording, cases where there is a wilful breach of rules, partly misunderstanding and partly use of discretion. We do not know the balance of the sources of under-reporting, but we know the overall shape of what has happened over the last 25 years.
Q140 Chair: I think you have answered my follow-up, which was: what is the difference between under-recording and misrecording? One is wilful and one is use of discretion or mistakes. We are interested in the relationship between misrecording and under‑recording.
Professor Hough: It is really hard to answer that, because it really ends up being a question of interpretation of intention. It is whether somebody knew about the rules and wilfully broke them, did not understand the rules, or was making what they thought was a fair judgment.
Q141 Chair: How much confidence do you have in the general intention of police officers to record crime accurately?
Professor Hough: I think that the proportion of wilful breaking of the rules is probably quite small. I have no evidence to prove that. I found some of the earlier evidence about clear rule-breaking behaviour worrisome.
Q142 Chair: Your methodology would not actually pick that up; I am asking a question outside the scope of your expertise.
Professor Hough: If you put the Crime Survey against the recorded crime statistics, you cannot get the answers to that question. You are right.
Professor Shute: Neither have the inspections that have been conducted to date been able to quantify the level of dishonest manipulation of the recording process by police officers and police forces around the country.
Q143 Chair: Can you describe how CSAC has actually contributed to an improvement of this situation?
Professor Shute: Part of our brief obviously is to help to underpin public confidence in crime data and the integrity of crime data. The main way that we have done that thus far is, as I have indicated earlier, by working with HMIC in helping to advise on the methodology that it might use, in suggesting that there ought to be systematic, regular programmes of external audit and scrutiny of what forces do, and that HMIC is the right body to do that—there is no other body that is able to perform that function—and to ensure that the methodology that they propose for future inspection is robust and fit for purpose. That is the way that the committee seeks to underpin that process.
Professor Hough: One of the key things we did, starting two years ago, was to shine some light on the divergence between the BCS/CSEW trend and the recorded crime stats. We have been trying to think how one might explain that over the last two years in discussion with HMIC and ONS.
Professor Shute: You may have had access to the methodological note that was published by ONS. If you do not, I would be happy to provide the Committee with a copy. It analyses the variations in crime trends by studying comparable crime categories between the Crime Survey for England and Wales and police recorded crime. It contains a foreword from the Chair of the Crime Statistics Advisory Committee.
Q144 Chair: How fair is it to say that you are interested in process, methodology and measurement, and you are not really qualified to judge about the leadership, the application of ethical practices, training of officers and that sort of thing? How fair is it to say you are not really qualified to judge on how much that part of the picture is distorting the statistics?
Professor Shute: We are able to comment on some of the things that you just mentioned. We are in a position to comment on governance structures that surround the collection and presentation of crime statistics. For example, if it appeared to us that there was inappropriate leadership in forces and that the wrong messages were being transmitted to police officers by Chief Constables and their senior teams about the significance of crime data and why it matters that it is reliable and robust, that would be an issue for us.
Q145 Chair: Can I ask you that question, then? What do you see that causes you concern, in terms of the messages sent by leadership, governance or even Police and Crime Commissioners?
Professor Shute: We have seen a number of things. We have seen the evidence that has been given to this Committee. We have seen the reports that have come out from HMIC. We have seen, for example, suggestions that the recording of crime by police is more robust where Force Crime Registrars are working efficiently and able to provide independent advice.
Chair: I am going to stop you there, because we are going to come on to that. Do you have anything else to add, Professor Hough?
Professor Hough: The impact of numerical targets on police statistics is on the borders of our remit, but it is important. Numerical targets have had a marked effect on the way that crimes are recorded.
Q146 Priti Patel: I would just like to build upon the previous question the Chair put to you both. How has CSAC helped improve police recorded crime data at a local level? Effectively, what would that mean to local police forces and to members of the public who obviously use crime data to benchmark their own views on what is going on in their communities, in terms of policing and crime?
Professor Hough: We can only really operate at arm’s length and look at the governance structures that affect local crime recording. Stephen has talked about HMIC, and then there is the Force Crime Registrars and their relationship to the Home Office and so on. We can only really intervene at that level to make governance tighter. We do not have larger, more direct levers.
Q147 Priti Patel: What are those points of intervention?
Professor Hough: We can advise about HMIC’s inspections; we can advise about the role of Force Crime Registrars; we can talk about their relationship to the centre, whether that is the Home Office or ONS. Things of that sort fall within our remit in terms of offering advice, but we do not have an executive function.
Professor Shute: That is right: we do not have an executive function. We have done a number of things to try to bolster confidence in crime data. We can analyse trends and we can look to see if those disclose worrying aspects of the current system. There are some concerns that come out of those trends. We have just described the divergences between the two crime series. We can offer advice on the way that crime data is presented—and have done—and the classifications that are used to sub-divide the 1,400 notifiable offences into manageable categories. We can and have offered advice on how the data ought to be presented in a way that is comprehensible to users and to the public more generally, and we have done. For example, we have suggested that high-volume types of crime, such as shoplifting and bicycle theft, be extracted from larger categories and identified separately. Similarly, we suggested that rape ought to be identified separately from the category of sexual offences, so that one can more immediately focus on that. There are a number of things we can do, but there are obviously limitations on what we can do. We are an advice body; we have a set of terms of reference that set out those bodies that we advise—in effect the producers of crime statistics, such as the Home Office and ONS, and the inspectors and auditors, HMIC.
Q148 Kelvin Hopkins: You have suggested the manipulation of statistics by police forces is not great, and yet in a previous session, we heard that this was much more significant than you have suggested, particularly in the Met. That is one point for you to comment on. The other point is that with HMIC, the Kent Police Commissioner had asked for a report on her police force and had a specific report done. A previous report by the same organisation nationally has shown quite different results for Kent. Doing an intensive survey, they found quite different reports than they had previously. Is HMIC the right organisation?
Professor Shute: There are two parts to that question; if I may, I will take them separately. There is the question about the extent to which the police are manipulating the crime recording process, either to conform to externally imposed performance targets or to improve the way that they are perceived in the locality or within the force. That may come about through a variety of reasons. It may not solely be down to dishonest, deliberate and intentional misapplication of the counting rules by police officers. That may be part of it, but it may occur through other routes. It may be that the police officers do not fully understand the Home Office counting rules and therefore misapply them. They have not done that dishonestly; they just do not understand and do not have the skill to apply—
Q149 Kelvin Hopkins: We had specific techniques of manipulating results described to us.
Professor Shute: Yes. Not all of them were necessarily the product of dishonesty. Some of them might be used dishonestly, but some of them might arise from non‑dishonest misapplication of the rules, even within the techniques that have been described to you.
Professor Hough: We do not have a sight line on the quantum of dishonesty versus other, but both exist, and I would not dispute that for a moment.
Professor Shute: We know that some police officers are prosecuted successfully for failing to perform their duties appropriately in this area. The question is what exactly the level of that is. For the minute, that is something that we do not have a hard measure for.
You asked about HMIC and whether we think that it is an appropriate external scrutiny body for crime data integrity and the process of recording crime statistics by the police. There are always legitimate questions to ask about what the regime for external scrutiny and audit ought to be, and who is the most appropriate body. Of course, in this area we have had two external scrutiny bodies at work: the Audit Commission, which carried out that function for about four years, and now HMIC. It is fair to say that HMIC was slow to step into the gap created by the cessation of Audit Commission work in 2006-07. We had to wait until 2009 before we had the first HMIC report, and we have only had two significant published reports. There is more that HMIC can do in this area, and my committee is encouraging HMIC to do more, as I indicated in my response to the Chair’s question on this.
If you do not have HMIC doing this work, it is hard to think who else would do it. The Audit Commission is no more. You need an inspectorate that has standing, has authority and understands the area that they are looking at. They have to have the appropriate skill set, because otherwise they will not be able to get to the bottom of the matter and uncover what is really going on. You have an inspectorate now that is led, for the first time in its history, by a Chief Inspector who has not come from a police background; that is the first time that has happened since Chief Inspectors were created within HMIC. That should give the inspectorate more distance, perspective and independence as a result.
Professor Hough: I thought the Kent report was actually a pretty tough report. If I were Kent Police receiving that, I would think, “This is a hard report with difficult messages.” It made me feel confident about HMIC’s ability to do that work.
Q150 Alun Cairns: Professor Shute, specifically why do you think that the police appear to be recording a lower proportion of crime reported to them than previously?
Professor Shute: It depends what you mean by “than previously”. It is worth pointing out that concerns about the misrecording of crime by the police have been around since the 19th century. In 1921, the then Inspector of Constabulary, Sir Leonard Dunning, wrote in his annual report for HMIC that he was concerned that police officers were, as he put it, slewing over complaints in order to give the impression that their detection efforts were more effective than they were. If the Committee wants that report, I would be happy to provide it with the relevant quotation. There has been this concern for a long time.
Q151 Alun Cairns: Can I stop you there? I want to know specifically why ONS came out with quite a strong statement and conceded that there was obviously a trend. I want to know exactly why that is the case now, not—
Professor Shute: Can I just finish? Yes, they pointed out leadership issues in that report; they also pointed out—
Alun Cairns: I know what they said; I want to know what you think.
Professor Shute: I think that they were right. Unless, as I was indicating earlier, the leadership makes it clear that crime recording is a key, central, core function for the police, and unless those messages go out strong, clear and loud, and are regularly transmitted, it will be inevitable that you will find regression from officers lower down in the hierarchy. Leadership, as with any organisation, is absolutely key. Structures, processes and systems are really important too, as well as the right leadership messages. I know that you are going to come to those issues in a moment, but Force Crime Registrars and the governance structures matter. In my view, there are improvements that can be made there, and I can set them out.
Q152 Alun Cairns: Can I interpret all that answer as lack of leadership and lack of scope? I am trying to be precise.
Professor Shute: There needs to be improvement in both those areas. In effect, structures, systems, processes, lack of ownership of this issue and lack of leadership results in decay in the recording systems.
Professor Hough: Can I say why I think the two measures diverged from 2007 and 2008? There were two or three critical things that happened: the Audit Commission stopped its audits; Sir Ronnie Flanagan’s review of red tape said the police had too much red tape; Jan Berry’s review came to much the same conclusion; and in 2008, KPIs were thinned out. Those things all signalled to the police that “recording is not quite as important as we thought it was in 2002”. I expected then there to be a divergence, and I am pleased to say it is one of the few predictions I have made that held up.
Q153 Alun Cairns: Exactly what needs to happen to gain consistency in the system?
Professor Hough: One needs to decide whether we want around 90% of crimes recorded. Is that degree of tightness in the recording standard right? If so, to get back to 90%, we need audits from HMIC and we need signals that, actually, what we thought was red tape is proper accountability. There are choices to be made about how fully the crime recording system works.
Q154 Alun Cairns: In considering under-recording or misrecording, and you talked about that to the Chairman earlier, would you say wilful blindness was part of it?
Professor Hough: I think some of the pressures of performance targets have made people wilfully blind, which is slightly different from explicit misconduct.
Q155 Alun Cairns: Was that a yes?
Professor Hough: That is a complicated yes.
Professor Shute: I have no doubt that there has been dishonest manipulation at one end, through wilful blindness, to misunderstanding and ignorance, to the inappropriate exercise of discretion within a complicated set of rules. The Home Office counting rules extend to 400 pages. They need to be translated in a way that allows police officers on the ground to use them appropriately, and that is not straightforward. All those things have combined together.
Q156 Alun Cairns: Each of you has mentioned the cessation of regular independent audits. What contribution did the withdrawal of that make to the greater divergence in terms of the crime data?
Professor Shute: As has just been indicated, it is one of the explanations for why the two series started to pull apart.
Professor Hough: We cannot quantify it. We cannot say it is 50% stopping of audits and 50% the signals about red tape and bureaucracy, but they were all factors.
Professor Shute: It is also the case that in 2000 and up to 2002, when the National Crime Recording Standards were created, wrapped around the Home Office counting rules, there was a consensus across police forces in the country that something needed to be done. The state of affairs at that point could not continue. Over time, though, that eroded. It was supported by the Audit Commission, although arguably somewhat undermined by the fact the Audit Commission described a 90% recording rate as good and a 95% recording rate as excellent. That is how their standards worked in this area. Nonetheless, they were out there busily reinforcing the messages at a time there was a consensus. That eroded over time, and that was one of the things that led to this divergence.
Q157 Alun Cairns: What recommendations will CSAC be making to bring about changes in consistency because of the concerns expressed?
Professor Shute: I think what needs to be done is a review of the governance structures around this, including the role of Force Crime Registrars and Force Incident Registrars. At the moment there is no generic common job description for those roles within each force. Although there is some indication of what their role might be within the National Crime Recording Standards, it is flimsy and in my view needs to be fleshed out into the form of a common job description. Force Crime Registrars need to have the authority within forces to ensure compliance. That means they need appropriate grades within the force in order to be able to command that respect.
There is a lot of evidence out there now to suggest that if Force Crime Registrars are effective, you are going to get greater compliance; you are, therefore, going to get higher levels of crime recording. They possibly need dotted-line reporting lines into bodies outside their own forces, perhaps into the National Crime Registrar or into the National Statistician. They need updated training and possibly accreditation, possibly carried out by the National Policing College, to ensure that they have the skill set that they need. You need to ensure that they are separated out from operational line management, and that there are not too many tiers of reporting between Chief Officers in forces and the Force Crime Registrars. A review of all that, plus more robust work by HMIC on a regular basis, giving it the external perspective, should go towards improving the position.
Q158 Paul Flynn: As the Police and Crime Commissioners are elected and will need to seek re-election, what is the difference in the pressure on Police and Crime Commissioners to fiddle the figures, rather than the Chief Constables fiddling the figures?
Professor Shute: I think they will have similar pressures, especially if they are setting numeric performance targets for their forces.
Q159 Paul Flynn: When you say “similar”, could you make it clearer? Who is under the greater pressure?
Professor Shute: We have seen pressures on politicians to misrepresent crime data over recent years. One of the reasons for taking the presentation of crime data out of the Home Office and placing it with an independent body, the Office for National Statistics, was to insulate the production of crime statistics from the political imperative.
Q160 Paul Flynn: This is a very interesting answer that you are giving me, Mr Shute, but could you answer the question? Who is under greater pressure: an elected politician who is a Crime Commissioner or an unelected Chief Constable?
Professor Shute: I think both are under pressure.
Q161 Paul Flynn: You have said that. Is there any difference between the two?
Professor Shute: They are in different places, but both have pressures to manipulate crime data to their own ends. We have, as a committee, issued guidance to PCCs, to their analysts and to their media departments about the proper way to present crime data to ensure—
Q162 Paul Flynn: You are clearly not going to answer the question. It is a very simple question. Is there any difference in the pressure on unelected and elected persons? The elected person has to appear before the public in a couple of years’ time, and if they appear before the public and say, “I have managed to increase crime by x%”, they are unlikely to be re-elected. Is that not true?
Professor Hough: It is true.
Paul Flynn: Would you not agree that the Police and Crime Commissioners are under greater pressure to falsify the figures than the Chief Constables?
Professor Hough: They are under greater pressure to do so, but they have fewer levers to make it happen, because they cannot penetrate through the—
Paul Flynn: Why is that? What levers do they have?
Professor Hough: They do not have the same direct levers of command and control as the police forces.
Q163 Paul Flynn: You wrote to all the candidates in the Police and Crime Commissioners election. You then wrote to all the elected Crime Commissioners. Why?
Professor Shute: Because we were aware that in the lead-up to elections, in the construction of manifestos, there might be a temptation on the part of aspirant PCCs to present crime data in a way that was inappropriate, and that that might lead to a decline in public confidence in crime data generally. That has significant disadvantages, and we wanted to guard against that by informing PCCs about what steps they needed to take to ensure that they presented crime data appropriately.
Q164 Paul Flynn: In the year ending March 2012 in Wales, a small homogenous country, one of the police forces reported an increase in crime of 3%, another reported a decrease of 3%, a third recorded a decrease of 7% and the fourth recorded a decrease of 15%. What interpretation would you place on that? Would you suggest that there is need for any action by anyone?
Professor Shute: That is exactly the sort of use of crime data that we were seeking to guard against. When somebody says that crime has gone down by 15% or increased by 15%, what do they mean? Do they mean that police recorded crime has declined or increased by 15%? Do they mean that crime as measured by the Crime Survey for England and Wales has declined or increased 15%? Unless they clarify that, it is a misleading statistic. It was exactly that sort of thing that we were trying to guard against.
Q165 Paul Flynn: Did you take any action when you saw those Welsh figures? Was the top one plausible? Was the bottom one plausible? Would you feel that you had to intervene when there was such a disparity between plus 3% and minus 15%?
Professor Shute: We would seek to take measures to ensure that those who use crime statistics know how to use them. We are not responding to individual examples of inappropriate use of crime statistics. That is not part of our function; that is not within our terms of reference. We are interested in improving understanding, and hopefully that will drive up the appropriate use of crime statistics by those who present them to the public.
Q166 Paul Flynn: What change has taken place in the reported crime figures since Police and Crime Commissioners have taken office?
Professor Hough: We do not know, in terms of under-reporting.
Q167 Paul Flynn: Do you know what changes have taken place in areas A, B and C around the country? Has there been a reported decline in the levels of crime or a reported increase?
Professor Hough: Overall, recorded crime has continued to fall.
Q168 Paul Flynn: Indeed. It has done for 30 years.
Professor Hough: I do not hold in my head the trends for individual forces.
Q169 Paul Flynn: If we take the last four years under the Chief Constables and the 12 months that we have had since, has there been a difference? Surely you should know that.
Professor Hough: It has been a consistent downward trend.
Q170 Paul Flynn: Has it continued the trend on the same level that has been happening for decades?
Professor Hough: We could obviously look at the stats and get back to you, but I do not know off the top of my head.
Q171 Paul Flynn: You have not seen any noticeable difference between the two in the last year.
Professor Hough: I have not looked at four years—
Chair: What difference would you expect to find?
Professor Hough: Over time, I really do not know if PCCs will respond to the pressures that they are under successfully to reduce recording. I do not know, but it is a risk.
Q172 Paul Flynn: There is a question at the moment of whether the PCC election will be the last, and there is a move to replace them. What is your view on the value of the PCCs so far, as far as influencing crime statistics goes?
Professor Hough: My view is it really is too early to say. I know that is sitting on the fence, but it is too early to say.
Professor Shute: One needs some accountability mechanisms for the police. We have replaced our police authorities with PCCs. The PCCs are serving a number of functions, one of which is an important accountability function. That is separate from whether they will be effective, through their very presence, in driving down crime within their areas. There are a large number of PCCs; they all are going about their work in different ways. I agree with my colleague that it is too early really to carry out a thorough review of their efficacy. I would not want to come to a definitive judgment so soon after their election and their creation.
Q173 Paul Flynn: Have crime figures been fiddled by a large group of police forces in the last five years, in your view?
Professor Shute: I have tried to answer that by saying that we do not know the extent to which fiddling has gone on. We do know that there has been quite widespread under‑recording and misrecording.
Q174 Paul Flynn: There is no accurate, reliable measure for deciding whether they were fiddled or not. That is what your evidence is this morning.
Professor Hough: We cannot say that.
Paul Flynn: You do not know.
Professor Hough: I do not know how much misrecording there is, as opposed to just under‑recording.
Paul Flynn: We hope for a bolt of lightning from above that will reveal what the truth is on this.
Chair: From HMIC.
Paul Flynn: Yes, there must be someone who knows.
Q175 Chair: On this question of quantifying the error rate, PC James Patrick and his fellow witnesses all seemed to agree it could be around 10%. How reasonable do you think that estimate is?
Professor Hough: Of misrecording?
Chair: Misrecording or just under-recording.
Professor Hough: We know in broad terms how much under-recording there is, and it is now roughly 70%. 70% of reported crimes get into police statistics. I do not think they know, and I certainly do not know what the balance is with misrecording.
Q176 Chair: Let me ask the question again. They said that of the crimes walking in through the police station door, so to speak, up to 10% of those incidents were being misrecorded or under-recorded. Do you think that is a reasonable estimate?
Professor Hough: I think it is an underestimate if it is misrecording and under‑reporting, because we know that it is a larger amount than that.
Q177 Chair: What estimate would you make of the percentage of crimes that walk in through the police station door that are not recorded correctly, for one reason or another?
Professor Hough: I think from the ONS paper it would appear to be about 70%.
Q178 Chair: 70% are misrecorded?
Professor Hough: No, 70% of reported go unrecorded. I can only repeat that I do not know the balance within that.
Q179 Chair: I think you are answering a different question. When something starts getting entered on to the Police National Computer as a crime, and then finishes up being nodded, cuffed or whatever, they are saying that 10% finish up being misrecorded one way or another. Do you think that is a reasonable estimate? What is your estimate? If it really is that you do not know, just say you do not know.
Professor Hough: I have said I do not know several times.
Professor Shute: We have no accurate handle on exactly what the proportions of wilful, deliberate misrecording are, but those numbers are possible.
Q180 Chair: How reasonable is it, therefore, for us to believe our witnesses?
Professor Hough: I think you need to ask them where the 10% figure comes from, whether it is a guesstimate or whether it is actually statistically based. That is all I can say.
Q181 Robert Halfon: When you answered my colleague Mr Cairns’ questions about recording statistics, you said some is genuine malpractice, some is by accident and so on. If you had a set of scales, how much weight would you give to the malpractice side of things, compared to the other things that you mentioned?
Professor Shute: Weighing them against what criteria? One case of deliberate, wilful, dishonest misrecording by a police officer is a very serious matter. The very people who are meant to enforce the rules in this country ought not to be breaking the rules when it comes to the recording of crime, so one case is a very serious matter. If it is a widespread practice, that is an even more serious matter.
Q182 Robert Halfon: You said that when they were recording statistics, some were malpractice, some were because of Government targets, some were genuine errors and so on. I am asking you, if you had a set of scales in front of you, what percentage you would put in the malpractice section of the reporting.
Professor Shute: If I have interpreted it correctly, that is exactly the same question as the Chair was asking me a moment ago, in which I indicated that we do not have good statistical information to answer that question. We have some impressionistic views, and you have had evidence before you along those lines. If the other witnesses have some research study or a way of grounding that in reliable statistical data, they should present it to you. We do not have it and we do not know of its existence.
Q183 Robert Halfon: What is your instinct? Do you think there is a preponderance, or do you think it is for the other reasons that you set out?
Professor Hough: My hunch is that, in misrecording, malpractice is a minority. That is a hunch, but that is where I am.
Professor Shute: My hunch would be the same. These are complicated rules for the recording of crime. As I said, it is a 400-page manual. It requires interpretation and translation. There is a deficit of expertise in forces to provide that translation. Therefore, there are systemic reasons for thinking there is quite a lot of misinterpretation and mistake‑driven activity. I would expect that that outstrips deliberate, wilful dishonesty. It is a guess, though.
Q184 Chair: To what extent is there knowledge of that accidental recording and then blindness towards it, because actually it is rather convenient for crime statistics to be recorded a bit lower than they actually are?
Professor Shute: I am sure there is some of that. For busy front-line officers and in busy control rooms, there is also perhaps not sufficient attention paid to the niceties of Home Office counting rules and the National Crime Recording Standards. They have other things that they feel deserve more of their attention. That is all about some of the issues we came to earlier—leadership and the right messages being transmitted down throughout the force. There is a whole variety of pressures.
Q185 Paul Flynn: It is crime statistics on which police promotions, decisions on massive spending, and whole policies of Government are based, and you are telling us that all the information we have is at best impressionistic. We have no clear guide to the accuracy of crime statistics provided by the police.
Professor Shute: I do not think I said that. What I did say was we do not know how much of the under-recording and, particularly, the misrecording is down to wilful, dishonest activity by police officers or, indeed, wilful blindness, rather than simply errors or failures to interpret the rules appropriately.
Q186 Paul Flynn: Are you happy with that situation?
Professor Shute: I am absolutely not happy if crime statistics are used inappropriately and if we cannot rely on them. For all the reasons that you said, and a host of other reasons, it is absolutely central that crime statistics in this country be trustworthy and command confidence, so that they can form the basis for proper decisions about how resources are allocated and so on.
Q187 Robert Halfon: Can I ask you a general question about the British Crime Survey? Am I right in saying that it does not include crimes against young people? Is that right?
Professor Hough: It includes people 16 and over in the main sample, and then there is a subsidiary sample of people 12 to 15, up to the age of 16.
Q188 Robert Halfon: What would you say is the flaw in that survey? People always say that it is not accurate and does not properly reflect—
Professor Hough: It is not precise, but also its coverage is not comprehensive, in that it does not include crimes against businesses, organisations and people in prison. There are large numbers of gaps, but it is a robust indicator of how much crime there is against people and their private property.
Professor Shute: It is also the case that the sample is not sufficiently large to allow one to disaggregate it by force area.
Q189 Robert Halfon: Who actually does it? You just see it reported. Who is responsible for the British Crime Survey?
Professor Hough: ONS oversee it. They contract the fieldwork to a company who does the interviewing, and analysis comes back in-house, largely.
Q190 Robert Halfon: Would you make recommendations for changing it and making it better, to encompass some of the things that you just highlighted?
Professor Hough: Its scope can be extended. For example, cyber crime needs to be captured more fully. There is work being done, but it could be done more fully. If there were buckets of money, I would want estimates that were viable at individual police-force level. At the moment, though, I cannot see that happening. It is just too expensive.
Q191 Robert Halfon: Whenever it comes out, if it shows a crime going down, people say there are too many flaws in the survey so it is not really worth it. Do you think it does give a useful snapshot?
Professor Hough: I think it gives a good snapshot and a better indicator of trends. If I had to choose—and I am obviously parti pris, because I have been involved in the Crime Survey for a long time—its trend is more reliable than the trend in police statistics. On what is happening now, I would say the police are over-stating the rate of decrease in crime.
Q192 Robert Halfon: Why is the trend more reliable?
Professor Hough: Because the methods of collecting are not affected by reporting changes and recording changes. What we have seen in the years from 2007 onwards are changes in recording practice.
Professor Shute: Key performance indicators for the police can affect the recording of crime by the police, but the Crime Survey for England and Wales will not be affected by that. We are fortunate in this country, in that we have a variety of sources for the data on crime and we can look at trends over time. Not all countries—arguably no other country—are in as good a position as we are to try to estimate what is going on. That is so valuable for all the reasons that were indicated a moment ago.
Q193 Robert Halfon: Are you saying that the more reliable indicator for crime statistics is the British Crime Survey, and that is what people should look at, rather than the police recorded crime?
Professor Hough: I would say it is definitely more reliable. It is less precise; it gives estimates with errors, but it is a better indicator of what the underlying trend is for crimes against people and their properties.
Q194 Robert Halfon: It is more strategic.
Professor Hough: Yes, it is broad-view. You have to spend a lot of money to get the narrow, focused view.
Professor Shute: It covers crimes that people have experienced but they have never reported to the police. Police recorded crime is blind to that.
Robert Halfon: Because it does not include non-reported crime.
Professor Shute: Yes.
Q195 Robert Halfon: Does the Home Office share that view?
Professor Hough: You would have to ask them, but they have said you need the two measures to understand what is going on.
Q196 Robert Halfon: Do they use both measures equally, in terms of the way they analyse crime?
Professor Hough: If they want to look at the force level, the Crime Survey for England and Wales does not enable a force estimate. You have to look at what is going on at—
Professor Shute: The only way of rectifying that would be to have a very much larger sample and, of course, there is an added cost. The only other thing I would say is that the crime survey needs to be carried out regularly in order to allow trends over time to be measured. You need continuity of funding for the Crime Survey. If it were to be withdrawn, that would be a serious problem.
Q197 Robert Halfon: Does the Home Office measure both forms equally and look at both properly, or does it give more weight to one survey than to the other?
Professor Hough: I do not think I can answer. They present the findings together in one volume.
Q198 Robert Halfon: But in terms of their own analysis and in the development of policy formulation.
Professor Hough: I would hope they do a synthesis of the two, but it is a question for them.
Professor Shute: As far as I know, they do. I would be surprised if they were not giving each the weight that it deserves in the areas where they can inform policy and practice.
Q199 Robert Halfon: When the police do their own recorded crime statistics, do they look at the Crime Survey, give an equal weight and bring the two together in any way?
Professor Hough: I would hope they know what the national picture is from the Crime Survey, and interpret their own trends in the light of that.
Q200 Robert Halfon: From what you were saying earlier, because there is the difference, if the police come up with one thing and they see that the British Crime Survey comes up with a completely different thing, do they then look at it and try to rebalance what they have come up with?
Professor Shute: One would hope that they are looking at the documents being produced on the variation in crime trends to cast a light on police recording practices in their force.
Q201 Robert Halfon: Do they ensure that the British Crime Survey reflects their own analysis of statistics and the recorded crime statistics, in essence?
Professor Shute: If they do not, I would have thought they had a problem.
Q202 Robert Halfon: Do they? Given all the problems you and others are putting out, and given that you say that there is a divergence of crime stats from the two different statistic gatherings—the Crime Survey and the recorded statistics—that suggests that they do not give proper weight to the British Crime Survey when they gather together their own statistics.
Professor Hough: I do not know if they have examined under-recording at the force level, but that is what the divergence would prompt me to do.
Q203 Chair: I just have one final question. You have been very helpful, and I have personally really enjoyed your very frank evidence. Thank you very much. If you were advising the National Statistician or the UK Statistics Authority on whether recorded crime should enjoy the designation of national statistics, what would your advice be?
Professor Hough: Yes, it is important.
Q204 Chair: Obviously they are important, but are they accurate enough?
Professor Shute: We are not the assessment body. They have to make that assessment. If they conclude that they are not accurate enough and they are not worthy of that kitemark, it should be removed.
Q205 Chair: I was asking what your view is.
Professor Shute: I think it would be a very sad state of affairs if we ended up in that position.
Q206 Chair: What is your view? How would you advise the National Statistician? In your view, are these statistics reliable enough to enjoy the designation—the royal seal of approval of national statistics?
Professor Shute: I would not wish to pre-empt their careful evaluation of that issue against the relevant criteria. You have to go through that process. It would be a mistake to state ex ante what an ex post conclusion would be.
Professor Hough: If they were downgraded, the system would collapse. We would not have—
Chair: That is not a reason for not withdrawing the designation.
Professor Shute: Of course not. The designation should be withdrawn if, as measured against the criteria, they do not meet the requirements for national statistics.
Q207 Chair: In your view, do they meet the criteria?
Professor Shute: My view is that it would be wrong for me, ex ante, to come to a conclusion. You have to go through that process.
Q208 Chair: I am bound to say, having thanked you for your evidence—Professor Hough, do you have anything to add?
Professor Hough: I would be sorry to see them lose their status.
Chair: I think we all would, but I note that you have not given them a ringing endorsement. You have not said, “Of course not”, so I will leave it at that.
Q209 Kelvin Hopkins: There is one last question from me. Professor Shute, you recommended a return to regular annual external audits and robust regimes. The concern I have is that the inspection organisation, HMIC or whichever it might be, might be too cosy with the police. Where we have seen inspections or regulators that are too cosy with the organisation they are supposed to be regulating or inspecting, we have problems. How would you ensure that it is a robust regime and a regime that also leads in the end to putting victims as the primary concern?
Professor Shute: Yes, absolutely. There is always that fear with inspectorates that they become too close to the inspected bodies. Sometimes academics refer to that as capture: they become captured by the bodies that they inspect. At the same time, of course, they need the relevant expertise, and if they are too distant from the practice they cannot turn over the stones because they do not know where to look. You always have to get that balance right. If you ask for a personal view, I think it is a welcome development that the Chief Inspector for the first time in the history of the inspectorate is now drawn from somebody who has not been a serving police officer in the past. You also have currently, on the Inspectorate, inspectors who similarly do not have a police background. If you go back a number of years, that was not the case. The methodology needs to be right. The Crime Statistics Advisory Committee is in a position to offer advice on methodology. It is part of our function and remit, and we take that very seriously. If, over time, an inspectorate shows itself incapable of exercising the degree of independence and robust scrutiny that is required, then that is a real problem for an inspectorate and you have to look at whether it is delivering what is asked of it.
Chair: We must draw a line. Thank you very much indeed for your evidence. We look forward to our next witness. Thank you very much.
Witness: Chief Constable Jeff Farrar, National Lead on Crime Statistics, acting on behalf of the Association of Chief Police Officers, gave evidence.
Q210 Chair: Welcome to this second session this morning on crime statistics. I wonder if our new witness could identify himself for the record, please.
Jeff Farrar: Thank you, Chair. I am the Chief Constable of Gwent Police and I am the National Policing Lead on Crime Statistics. I am also a member of CSAC.
Q211 Chair: Could I jump straight into your evidence, in which you have said that there is little leeway for application of professional judgment in deciding whether to record an incident as a crime or not? You go on to say, “This often brings Crime Registrars and their staff into direct confrontation with police officers who perceive them to be prioritising compliance with the rules over the needs of victims and the wider public”. Can you explain what you mean by that?
Jeff Farrar: Chair, I think there are a number of points about why crime would not be recorded in compliance with Home Office counting rules. Public interest might be one, but there are a number of other issues around IT systems, performance pressures, lack of competence and training of officers, failure to do their duty, and just human error.
There are lots of reasons why stuff might not be recorded properly, but let me give you one example. It is one from my own force, where a member of the public finds their 11-year-old stealing from their purse. They phone the police for some help; they want some service; they want their child to be told that is not the right thing to do and to remember that for the rest of their lives. This is a daily occurrence for the police. This is the type of thing that we do frequently, and there are a number of examples. A police officer turns up, speaks to the child, gives them something to remember and hopefully puts them on the right path for the rest of their life. That is a crime and in the Home Office counting rules should be recorded as a crime, and there should be a crime outcome. That crime outcome could be a caution; it could be prosecution; it could be community resolution, but there should be a crime outcome. That then goes into the system. The reality is: would that member of the public have phoned us if they thought that was the approach we were going to take?
These may be a very small number of incidents, but I know from an audit that I have done in my own force that those incidents where we have failed to comply with Home Office counting rules have been for two reasons. One is because they genuinely thought they were helping the public to deliver a service. The second is that they had a lack of knowledge or understanding of Home Office counting rules. That was the sole reason for those failures. There are lots of reasons why it can happen, but it is really important that we allow officers to use common sense, albeit there has to be a line and we still need to make sure that they are complying with Home Office counting rules.
Q212 Chair: I have to say, Chief Constable, that this is a very novel piece of evidence you are giving us, because nobody else has suggested that there is any significant under-recording of crime at the victim’s request. On the contrary, the evidence is completely the opposite. I find this assertion that there is any statistically significant under‑recording of crime for the reasons you have suggested to be rather difficult to believe. How statistically significant do you think it is that crime is under-recorded for these reasons?
Jeff Farrar: I think it is probably on the margins. It is not the main amount of crime that will come into us, but it does stop officers doing what they are there to do, which is deliver the service to the public. If I can explain how I have compiled the written evidence: this is the written evidence from all 43 forces. They were all asked to comment. The evidence I present in my written submission is not just the evidence of Jeff Farrar sitting here. It is from the 43 forces in England and Wales, and the submissions that have come back to me. That is a recurring theme around service as opposed to compliance.
Q213 Chair: Can I quote Mick Creedon, the Chief Constable of Derbyshire? He said, “My fear is that inadvertently we are all still putting pressure on officers to do all they can to manipulate and create crime reductions”. That is rather at odds with your evidence. I do not think your evidence reflects Chief Constable Creedon’s view.
Jeff Farrar: The point I talk about on use of discretion and public interest is just one part of many reasons why crime might not have been recorded properly. I was there when Mick Creedon made those points, and the points he made are based on his view of what he has seen. I would say that nobody joins the police service to record crime inappropriately. They just do not. They join to help the public. If there is a pressure to misrecord crime, the question has to be: is that driven by targets or numbers? If it is driven by targets and numbers, are we asking people to do the wrong things? That is a question for the service because there is an inconsistency across the service in how that is being done.
Q214 Chair: What is the role of the national policing lead on crime statistics?
Jeff Farrar: It is a really interesting one, because I have been doing this role for the last 12 months, and it is not my day job. I am a Chief Constable in a force in England and Wales. I am not paid any extra money for doing the role; I am not required to do the role. I do it because I would like to get greater consistency across crime statistics in the UK. I think if the one thing I achieve in this role is greater consistency, then I will have achieved something. That is what I see as my role: it is getting a concerted effort across the service to get a picture of what is happening across the 43 forces.
Q215 Chair: How consistent do you think the understanding amongst chief officers is that there are risks to data integrity posed by the target cultures under which police forces have been operating for a couple of decades now?
Jeff Farrar: Every force in the country will tell you there is a risk of a target culture driving the wrong behaviour. That is inevitable. The work that HMIC will do and has done in the past has shown us that is absolutely clear. The work that they will do next year on the crime recording reviews will be hugely helpful. If you ask me that question, I run Gwent Police through a very ethical and value-driven approach, but on a quarterly basis there will be, for want of a better term, a league table of numbers that will be produced. It is human behaviour that you are inevitably going to look at where your force is in that list.
The reality—and in the conversations I have had with the Police and Crime Commissioner we are like-minded on this—is that we absolutely need to be driven towards increased quality of service for the people of Gwent, and that needs to be our overriding driving issue rather than following numbers. My experience, certainly from going to public meetings, is that the public do not want to hear numbers; they want to hear how we are delivering the service to them. That is really important. Once the service is judged, whether that be through ministerial direction, the media or other sources based on numbers, there is a real danger that you can end up driving the wrong culture.
Q216 Chair: You refer in your evidence to the role of FCRs—which, if I remember correctly, stands for Force Crime Registrars—and we have heard how inconsistent their role is across different forces. What is ACPO doing about this?
Jeff Farrar: I have been around the whole country and I have seen every single Force Crime Registrar, bar in one region. One of the things that was stark to me was the difference in status, the difference in role—
Q217 Chair: What are you doing about it?
Jeff Farrar: I have already met with the national policing college, and they are now starting to put in place the national training and accreditation for Force Crime Registrars. The stuff that comes out from the HMIC inspection will be hugely helpful in directing that. I have already written to every Chief Constable in the country about the engagement they have with their FCR, the role of their FCR and the status that they should have in the organisation.
Q218 Chair: What are your top three recommendations to all Chief Constables?
Jeff Farrar: I jotted a couple down before we came in. The code of ethics and what that drives us towards is very important for the leadership of the service. Training, not only for Force Crime Registrars—that training will be set up nationally—but for front-line staff in understanding the Home Office counting rules is very, very important. Consistency of IT and ensuring that that is capturing crime recording are essential for us.
On the status and role of FCRs, there is a real danger if you put someone in that role who is an in-service officer who is aspiring to go to a higher rank, the chance that they will have an open door to their chief officer is more problematic. It is not insurmountable, because it does work well in some areas, but my recommendation would be that, to get consistency on that and independence, it is difficult to put an aspiring sergeant or inspector in that role. I have two final points. The first is that internal scrutiny and audit needs to be much clearer across the service.
Chair: Is this about FCRs?
Jeff Farrar: No, I think—
Chair: I would be grateful if you stuck to the question I am asking. I asked for your top three recommendations about what Chief Constables should do about their status and their relationships with FCRs.
Jeff Farrar: I think they need to play a really important part in scrutiny and internal audit. Some already do, but it is not consistent. Their training needs to be national and accredited, which we have already lined up. They need an open door to a chief officer to sign off crime statistics—that is not a chief officer who has a responsibility for the operational area of reducing crime because there can be a difference—
Q219 Chair: What rank is an FCR?
Jeff Farrar: They vary. In some forces—
Q220 Chair: What rank should they be?
Jeff Farrar: I do not think they should be a police officer, personally, but they are police officers in many forces.
Q136 Chair: What rank should that police officer have?
Jeff Farrar: I do not think it matters what rank they have, as long as they have the access to a chief officer and the right status within the organisation.
Q221 Chair: So it would be okay for a sergeant to do it.
Jeff Farrar: Sergeants do do it.
Q222 Chair: That is okay, is it?
Jeff Farrar: Some do it very well.
Q223 Chair: Do you think a sergeant has sufficient authority throughout a large constabulary to create the right atmosphere around the authority of these guidelines and procedures.
Jeff Farrar: That is what I started with. I do not think a police officer should be in that role, personally. I think with an aspiring police officer, it is very difficult for them to identify failings. It becomes harder. It does work well in some forces around the country that I have been to. They do have sergeants who perform that role very effectively, but it is an inherent risk if you are going to put a sergeant or inspector in that role as opposed to somebody who is more independent.
Q224 Kelvin Hopkins: You mentioned ethics in your previous question and the College of Policing is proposing a new code of ethics. How do you think that could be used specifically to improve accurate crime recording? Will this reach right down to the police constable on the beat?
Jeff Farrar: It is a really welcome step, but it is not something completely new, in that we have had a statement of common purpose for the last 20 years; we have had our oath that we swear when we join office. It all talks around integrity, standards and what is expected in the service. I do think it would be a very positive step around setting the standards that are required for the service. It is going to be underpinned by legislation and the reality is it will be a benchmark for which we should be judged as a service. It is a really positive step for us.
Q225 Kelvin Hopkins: Will there not be something of a conflict at times with the natural solidarity of police forces? They have to work together in sometimes dangerous situations. The lads are going to be loyal to each other. At the same time, they have to behave and report ethically. How far down the chain of command will your concern about ethics go? Will it be on the bobby on the beat?
Jeff Farrar: It is about leadership. The vast majority of people who serve us as police officers join because they care about their communities; they are ethical; they are fair; they have integrity. We will always get the odd individual who does not aspire to that or does not comply with that. My experience of people I have worked with for more than 30 years in the police service is they join the service for the right reasons. I think it is about strong leadership; it is about ensuring people understand that code of ethics, that it is clear and it is driven by the values of the service and by individual services. It needs to be a value-driven organisation. If that is not in place then you are always going to have problems.
Q226 Kelvin Hopkins: The other conflict is within targets. Targets inevitably are a concern of the leadership. If the leadership is saying, “We’ve got to get up that table of performance in the country. If we report too many of our crimes we are going to go down rather than up,” in a sense, are you going see police constables saying, “I don’t care where we are on that table. We are going to report every crime and make sure the public knows that we are being truthful”?
Jeff Farrar: I think that is the evidence around the country already. If you look at the evidence you have had presented to the Committee already, there are people who want to do the job and serve the public. Again, it is back to my question. Nobody joins the service intent on misreporting crime statistics, so why would it happen? Yes, if you have an arbitrary target, inevitably if people feel they are going to be judged by not reaching that target then you can end up with unintended consequences. For example, I am working as an inspector in a locality; I am being asked to reduce bicycle theft by 10%. This month it looks like I am going to reduce it by 18%. What do I do? Do I tell people to slow down because next month I will be judged again? If I am only reaching 6%, what do I do, if I have tried everything ethically to reduce that crime appropriately?
The message I send in my own force is it is not about the numbers; it is about the service that we actually provide and that is what is important. I think there is inevitability, though. Even if you do not have a superintendant or a chief officer who is saying, “I want you to reach that target”, it will get subtly translated in different ways as we go down the organisation. When the service is judged by an inspection process that uses a league table or when it is judged by a media headline, there is inevitability that it can drive the wrong behaviour. It takes somebody with strong moral fibre to stand up to that and say, “No, I am doing the right thing for our locality”. The service needs some help on that.
Q227 Kelvin Hopkins: The gap between the independent recorded crime and police recorded crime is substantial. We have to narrow that gap, do we not? Do we not have to have Chief Constables in particular, and those at every level, saying that even if we do not pursue the crime to prosecution or whatever, it has to be recorded, and that every report to the police of a crime has to be recorded and be available for public inspection?
Jeff Farrar: It is a really interesting point, because America, which is five times the size of us, records only 50% of the amount of crime that we record in the UK. There is no other country in the world that records as much crime as the UK, yet would you say that we have more crime than elsewhere in the world? I do not think so, if you look at some of the other countries. Mike talked about it: we still know that 60% of crime does not get reported. Peter Fahy gave evidence to the Home Affairs Select Committee last week, or the week before, and he said asked whether, if we walk down the street and see somebody sneezing, that should be recorded by the health service as poor health. We are going to have crime in society that does not always get recorded. A really important point is, regardless of what it is recorded as, what service does that member of the public get when they report that to the police? That is a critical factor. My experience is that victims do not necessarily care how something is recorded; they care about what service they get. That is really important, and I wholeheartedly agree with the point you made in your question. The public deserve that, and they deserve nothing less. If we are not doing that, we are abrogating our responsibility.
Q228 Chair: Can I be clear that what we are talking about is not the kind of crime that is not recorded because it is not reported? What we are concerned about is crime that is reported and then for some reason or another is not recorded correctly. How do you have confidence that that 18% reduction in cycle thefts in your constabulary is accurate?
Jeff Farrar: In my force, for example, I have about 25 analysts. The question was asked earlier: what do we do with the data? They pore over that data because, actually, how do I employ resources effectively if I am not using that analysis effectively? If somebody reports to me that there is an 18% reduction in crime, my first question is, “Is that a reporting issue or are we reducing crime?” I will go into the detail underneath it to look at what the analysis says, look at what sorts of information we are getting in and what we are getting from our local PACT meeting about what people are saying about crime in their locality. That is the view my colleagues across ACPO. They would be doing exactly the same thing. If they are not, they are really not doing their job effectively.
Q229 Chair: When somebody reports a cycle theft, it is usually because they want to claim the insurance and, usually, the police officer will give a crime number so that they can go to the insurance company. Presumably there is a bit of pressure on your officers now to investigate these a little bit more, and perhaps to suggest to the victim that he left it somewhere or that a relative has borrowed it, or there is some other issue. How do you know that it is an 18% reduction in cycle theft?
Jeff Farrar: It is a really important point. It is my point about checking those figures and not just taking it on face value that there is an 18% reduction. It is looking at the reasons why.
Q230 Chair: How do you check the figures?
Jeff Farrar: Checking the figures—the first thing is: are we driven by the figures or are we driven by the service?
Q231Chair: How do you check the figures?
Jeff Farrar: I will check my crime-recording figures in a number of different ways. My Assistant Chief Constable runs an auditing group that has the Police and Crime Commissioner’s deputy on. There is that scrutiny.
Q232 Chair: We know there is an auditing group. Can you answer the question? How do you check the figures?
Jeff Farrar: I am not sure what you are asking me. I am saying I will do audit; I will do intelligent check on it.
Q233 Chair: How does the audit operate? What does the auditor do?
Jeff Farrar: For example, the FCR will routinely pick out incidents, particularly those that appear to be a crime on reporting and are not then recorded as a crime, to check the reasons why. Every force in the country has an element of audit around how they check that what is being reported as appearing to be a crime but is not being recorded as a crime is there. That is the stuff that is coming through as incidents.
Q234 Chair: Sorry; you have not answered the question. How do you check the accuracy of a single reported crime?
Jeff Farrar: That is the way it is done. It is done through the Force Crime Registrar.
Q235 Chair: I do not have a clue. You have not given me any idea of how you verify the truth of a cycle-theft statistic. How do you do it?
Jeff Farrar: The incident gets reported, either to a patrolling officer, to a police station or by phone.
Chair: Yes, that is the initial recording.
Jeff Farrar: It then comes into a computerised system, which will vary across forces. Within that system, there will be an auditing bureau, which is run by the Force Crime Registrar, that will audit and check those crimes.
Q236 Chair: What are they auditing?
Jeff Farrar: They are looking to see whether they are compliant with the Home Office counting rules and the National Crime Recording Standards.
Q237 Chair: What does that mean in terms of—
Jeff Farrar: If they are not in compliance with those rules—
Chair: How do they check how the rules have been complied with?
Jeff Farrar: They look at the incident; they look at the circumstances around it.
Q238 Chair: Whom do they talk to?
Jeff Farrar: They will talk to the officer in the case?
Q239 Chair: Ah, thank you. We are getting somewhere. What sort of conversation takes place?
Jeff Farrar: It will depend on the nature of the—
Chair: The cycle theft.
Jeff Farrar: It will depend on the nature of the crime. If that has been reported as a theft of a pedal cycle, within most forces they would have a crime management unit that would look at whether the investigative opportunities were exploited, so that everything was done that could be done to investigate that case to make sure that we detected that crime. There would be a conversation with that constable or sergeant to ensure that those lines of enquiry have been done. That is what will be done with the individual crime, but it will vary across forces.
Q240 Chair: In conversations I have had with various police officers since we initially took our evidence from James Patrick and others, I have been left in no doubt about what actually happens. A police officer will enter a crime on to the police system, and then there will be some arbitraging. How often do you find that a crime has been altered? When you look at the computer logs, how often do you find the crime has been changed?
Jeff Farrar: I can give you the details from my own force. Of crimes that are recorded, those that end up as no crime—becoming a crime and then being removed as a crime, so there is additional variable information that says, “This shouldn’t be a crime”—was 1.7% last year.
Q241 Chair: Across your force?
Jeff Farrar: That is in my own particular force.
Q242 Chair: That was to no crime.
Jeff Farrar: That was to no crime.
Q243 Chair: What about from cycle theft to lost property, or cycle theft to something else?
Jeff Farrar: I do not have the specific data with me. I would be more than happy to give you that in writing at a later stage. It is a very small number that go to no crime.
Chair: Or where a break-in is reduced to criminal damage?
Jeff Farrar: I think you are much more likely to have cases that do not get into the system in the first place, where we do not record them. For an officer then to have to record that as a no crime is a significant amount of checking.
Q244 Chair: When you are auditing, how do you find the crimes that were never put into the system in the first place? It is no good looking at the figures, because they are not there.
Jeff Farrar: That is the difficulty if you do not have them coming in. If they come in through your telephone system, you have them recorded as an incident. That incident can then be checked. If they report it to me on the streets or to my public protection team who interview a victim who discloses a number of sexual offences, or somebody walks into the police station and it is not recorded on our incident system, it is one of the things that HMIC has picked out on the forthcoming inspections as a real issue for us. How do you ensure that they come into the system in the first place? There is a risk there.
Q245 Chair: How much do you look at records of people who have been into the police station and then go back to the victim? How often do you go back to the victim of a crime to find out whether they feel the crime has been recorded properly?
Jeff Farrar: Interestingly, in the review that I have just conducted in my force, we did exactly that. Of 1,000 crimes over a 10-day period, we took 50 that appeared to have been crimes that had not been recorded as such. Twenty-five had not been recorded as such, and there was a genuine reason why, in line with national crime recording standards; for 25, there was not. We went back to the victims and said, “What sort of service did you have?” and well over 90% said they thought they had an excellent service. We do do those samples across the service, but again, as I started by saying, it is inconsistent.
Q246 Chair: 10% of your sample, randomly selected I presume, said they had not had a good service and you had not recorded their crime.
Jeff Farrar: Yes.
Q247 Chair: That rather bears out what we have been told—that there is a problem—does it not?
Jeff Farrar: If we are saying, “Are we ever going to record all crime, 100%, to a perfect standard?”, I would put it back into the context of looking at how stuff is recorded—not just across crime recording, but other crime data within the UK, and in other countries, and the standard is still pretty tough.
Q248 Chair: Looking at your draft code of ethics—honesty and integrity, fairness, equality, diversity, lawfulness and all these others—when you find that 10% of your sample is complaining, bearing in mind you are telling me that for 50% of that sample, there should have been a crime recorded but there was not, but the public are not complaining about it, what conversation do you have with the officers concerned who misrecorded that crime?
Jeff Farrar: The context of that is it was 1,000, and we took 50, 25 of which were done properly, and 25 of which were not. They were public interest, like I said at the start of my evidence, and a lack of knowledge and understanding of the rules.
Q249 Chair: The same applies. What conversations do you have with the officers concerned who misrecorded that crime?
Jeff Farrar: If we identify shortcomings that are misconduct, we would deal with it under misconduct.
Q250 Chair: Is it necessarily misconduct?
Jeff Farrar: It might be a neglect of duty, yes.
Q251 Chair: It might be, but it is in the culture, is it not? This target culture has been around for a long time.
Jeff Farrar: We have had targets for some time, but I was in the service back in the late 1980s and 1990s, where we had limited performance management; we had no targets. If you had asked me the questions you are asking me today then, it would not have been as good as it is now. However, I really do think there is a place for performance management in the service. My own personal view: is there a place for targets? I think they can drive the wrong behaviour. There are not any targets in my own particular force. It is varied across the country. If you have a personal target, it will achieve things; if I want to run the marathon, I will have a personal target. Once you make those organisational targets, they can drive the wrong behaviour, and you have seen evidence that it does potentially drive the wrong behaviour. I think it is really important for us as a service to recognise that we do have to deliver the best possible service to the public. There has to be performance management around that.
Q252 Chair: How do you make it clear that, whatever aspirations you set in your policing policy in the force, the code of ethics is actually far more important? How do you make that clear?
Jeff Farrar: If you spoke to officers in my force, they would tell you that the one thing I put up on the wall is our purpose, which is to protect and reassure people, and our values. People talk to me about technical skill. If it is not done in a responsive way that is fair, if people do not trust you, and you are not caring, if you are not doing those things and using the values set out in the code of ethics—Alex has asked us all, going across the whole service, to sign up to those values—all the rest of it does not really matter. That is the code that we need to be held to, and I think that is really important.
Q253 Chair: How do you know that your officers are actually subscribing to those values and believe in them? I know quite a lot of policemen. There is quite a lot of cynicism, is there not?
Jeff Farrar: There can be.
Q254 Chair: How do you make sure that you have really inspired people with those values?
Jeff Farrar: I think you have to value that behaviour. Where are the examples in your service of valuing people who stood up to those values and that commitment? That has to be a strong part of the performance measurement, as is being an employer of choice, as opposed to judging people based on numbers. If I have an inspector who does not know whether their crime has gone up or down and does not know what they are doing about it, they are not doing their job. That is a very different thing to an inspector who is not meeting their 10% target this month. The subtleties get lost.
Q255 Kelvin Hopkins: I have two concerns. There is the question of violence, particularly sexual violence; writing down crimes, from rape to sexual assault or whatever, is a serious concern. That has been brought to our attention in previous sessions. Would one simple way of dealing with this be to have all first reports recorded, so whatever people say is recorded at one point, and then what you finally decide is going to be the crime is recorded separately? You might get 100 crimes reported a week, and all of those are recorded as they come in at the first contact; the second point might be that 75 of them proceed to be an offence. Then we would know how many of those alleged initial thoughts about crime are being reduced, and what the percentage is. At the moment we only get the 75 of them. Could we not have that different approach?
Jeff Farrar: When the National Crime Recording Standards were introduced in 2002, there was this question about prima facie, which was to take everything on face value, to balance the probabilities. It is not a universally held view amongst all my peers, but I think for certain types of offences, you have to have a prima facie case. If somebody says, “I’ve been raped”, we record it as a rape. We do not investigate that immediately and ask them why they said they had been raped, and look not to record it in the system. Certain types of offences, like rape, in my view, should be recorded as rape, and if there is additional variable information that says it is not a rape at a later stage, then it should be no crime. I do not think there should be any way around that in the service.
However, let me give you an example of criminal damage. My previous chief went out on patrol, and went to a house where the lady said, “I’d like to report criminal damage”. If we took that on prima facie, we would record it, it would go into the system, and there would be a lot of bureaucracy around it. The glass was on the outside. He said to her, “How did it get damaged, then?” “Well, I kicked the window myself in a temper, but if I don’t get a crime number, I can’t get it replaced by the local authority.” That is not a crime. It might be a crime by her.
Chair: It sounds like fraud to me.
Jeff Farrar: But it is not a crime in the genuine sense. For some types of offences, I think there is room for balance of probability. For certain types of offences, like the ones you have talked about, I think it should be a prima facie case, and we should just be recording it; we should not be asking those questions.
Q256 Paul Flynn: Can I first of all congratulate you on your appointment, and say that your distinguished work on Operation Jasmine is appreciated by all Gwent MPs? You might find the questions I am going to ask you challenging, but I really wish you well, for your sake and for the sake of the people of Gwent. Could we go into your appointment, first of all? The Police and Crime Commissioner, when he was interviewed on 3 July by the Home Affairs Committee, was asked about how many candidates were likely to apply for your job, in view of the fact that Carmel Napier was pressurised to resign, and she was the second Chief Constable who had been pressurised to resign in my period in this House. He said that he expected a large number of candidates to apply for the job. In the event, how many were there?
Jeff Farrar: There was only one candidate, but there are lots of reasons for that. One of the things across the service is if you have a strong internal candidate, there are often not a lot of applicants. There are other reasons, Mr Flynn. The first is that we know that Wales has never been popular, in terms of attracting chief officers to come into Wales, for lots of reasons. I think it is a time in the service when, if you look across the whole of the service, there is one applicant for many job across many forces, and there are lots of reasons for that. To read that into the PCC would probably be unfair.
Q257 Paul Flynn: I am aware this is not unique, but these were special circumstances. If we can go back into the very interesting story of crime figures in Gwent, I quoted earlier the year when one Welsh police force recorded a 3% increase and Gwent recorded a 15% reduction. Damian Green said in Westminster Hall that in the period from June 2010 to June 2013, crime in Gwent went down by 29%. In the past year since the election of the Police and Crime Commissioner, it has fallen by a further 4%. You were a serving officer during that period. Do you have any comment to make about those figures, and do you find them plausible now?
Jeff Farrar: Absolutely. You have to look at the historical trend, and where we have come from. Gwent Police went through a reorganisation in 2009 that changed everybody’s job, from the Deputy Chief Constable right down to PCSOs. It was a wholesale change.
Paul Flynn: Could I just say—
Jeff Farrar: It would be helpful for me to give you a little bit of context, because if you look at the historical change at that time, in 2009, Gwent was the only police force in the country that had seen an increase in crime. If you try and compare us with other Welsh forces, Dyfed Powys, for as long as I can remember, has always been the lowest recorded crime in the UK. It is top. At that time, Gwent Police was 42nd out of 43, so you are not comparing apples with apples. For us to reduce crime by 30%, we had a long way to go. I think other forces in Wales, particularly Dyfed Powys, did not have that distance to travel, so to expect them to make the same reductions as us would have been unfair. We are not comparing the same things.
Q259 Paul Flynn: You accept the figure that Gwent Police reduced crime by 29% in three years, and then by a further 4%. Do you find that entirely believable?
Jeff Farrar: I can tell you the work that went on in the force during that three‑year period and it—
Q260 Paul Flynn: Do you believe it?
Jeff Farrar: When you say, “Do you believe it”—
Paul Flynn: do you accept what the Police and Crime Commissioner said in his campaign—that he was concerned about possible inaccuracies in the crime figures? He has said it since. Are you in disagreement with the Police and Crime Commissioner because of those figures?
Jeff Farrar: The Crime Commissioner and I are absolutely like-minded about what we want to do in Gwent Police, and that is to provide the best possible service to the public. When he travelled around the county, in a lot of meetings, people were saying that their engagement with the police was not what they thought it should be. We have really turned that around. I do think he made a very valid point about engagement with the community, understanding community needs and delivering what they want. That is why I did the internal audit for the service, and it has brought up the issues I talked about earlier.
Q261 Paul Flynn: HMIC said the Gwent crime figures showed 120 incidents that they were concerned about, and gave “cause for concern”. Ian Johnston said, “I think that in Gwent we were so focused on reducing the recorded crime that we got away from providing the service for the people who pay the wages”. He was clearly unhappy with those figures.
Jeff Farrar: On the focus of the force that he talked about, and the priorities around quality of service to the people of Gwent, he is absolutely right. There has been a very strong shift in focus in the last six months in which I have been in post as Chief Constable. That issue around—
Q262 Paul Flynn: We are not talking about that. We are talking about the period when you were a serving officer in Gwent. There was an astronomical increase—the highest increase in the whole of England and Wales—the 15% figure. Was that the truth, or were those figures unreliable, inaccurate or fiddled?
Jeff Farrar: I have talked about the reliability of figures throughout.
Paul Flynn: I have some other questions to ask you on that.
Q263 Chair: May I help for a minute? Chief Constable, how could you demonstrate why we should have confidence in that very large crime reduction? Did you conduct an audit during that period? Was there an external audit of that figure?
Jeff Farrar: HMIC did an inspection during that 18-month period, but I would say—
Q264 Chair: What did it show?
Jeff Farrar: It showed that we are up there as good, in relation to compliance around crime recording. That is what was in their audit during that period.
Paul Flynn: But there—
Jeff Farrar: Paul, I know you know this, but there was a huge amount of incredibly positive, hard work by our staff in focusing on those people who were committing crime in Gwent, and in changing the way that we structured the way that we did that. A lot of that hard work was widely attributed to those crime reductions. Would there have been misrecording in there? Yes, undoubtedly. I have already said that there would be. Do I think it was wholesale manipulation of the figures? No, I really do not.
Q265 Paul Flynn: Why can you not tell the Committee what the HMIC said? They said, about the crimes, “17 had been wrongly closed without a crime being raised, which indicates that crime recording in Gwent gives some cause for concern”. They were unhappy about those figures. The candidate for Police and Crime Commissioner was unhappy, and you are saying that they are okay; these figures were alright.
Jeff Farrar: I am not saying they are okay, but if you look at those 17 and the reasons why they were misrecorded, there was no evidence in any of those that that was due to a culture that was driven through performance pressures. It was all to do with the stuff I have talked about, which is lack of knowledge and awareness, IT systems, public interest and all of that in there. It again broadens the reasons why that figure is never going to be 100% accurate.
Q266 Paul Flynn: Do you think the previous Chief Constable, Carmel Napier, did all that she could to guarantee the accuracy of those figures?
Jeff Farrar: I think Carmel Napier was a very hard-working Chief Constable.
Q267 Paul Flynn: Then why was she sacked?
Jeff Farrar: She wanted the right thing for the people of Gwent, and I think Carmel Napier and Ian Johnston have both given evidence to the Home Affairs Committee about the period of time when she was the Chief Constable and he was the PCC.
Q268 Paul Flynn: I was present at their examination. She did say, “We set up, under the Police Authority, a data quality and assurance group, which took the feedback from 2011 HMIC inspection of the force in relation to crime recording”. She said, “As Chief Constable, I put all the mechanisms I could possibly put in place to make sure that both our incident and crime and anti-social behaviour recordings were as accurate as they could be”. What more can you do now?
Jeff Farrar: You have already heard of what I have done. I have trained front-line staff in relation to knowledge of the Home Office counting rules; they all get trained on appointment in relation to knowledge of the law, so they have had specific training around Home Office counting rules and what the standards are. I have made it very clear to people what my expectation is around ethical crime recording and the standards I expect. My Force Crime Registrar has a direct link into me, and in the audit group that I have set up, I have included the PCC’s office; we have had a conversation around it, and you have seen the evidence of that from the report that is public today around the inspection in the force. I am open to any direction on what else I could do, but we have done everything we possibly can.
Q269 Paul Flynn: Were you acquainted with Ian Johnston before his election?
Jeff Farrar: He was a serving superintendent at the same time I was a serving superintendent.
Q270 Paul Flynn: If we look at the future on this and the past evidence we had, Mr Johnston is now a politician, not a serving policemen. To get re-elected, he has to do what all politicians do and please the electorate. He has already signed off on the fact that there was a 29% reduction in Gwent. Add to that another 4% reduction. Is the likelihood now not that if more accuracy is put into figures, it is likely to record an increase in crime? Is he not under pressure to avoid that?
Jeff Farrar: I am absolutely determined that crime within Gwent will be ethically recorded, and I will do everything I possibly can to do that. That is a conversation that I know is shared by my Chief Constable colleagues. If that means increased recording of crime, then so be it. The one thing I would say about the Police and Crime Commissioner is never once have I had a conversation with him about re-election, and never once have I had a conversation with him about numbers on crime—keeping them down. Every conversation I have had with the Police and Crime Commissioner has been about doing the right thing for the people of Gwent.
Q271 Paul Flynn: We have invited Ian Johnston to come to this Committee, but he was on leave at the time, as you are aware. You have seen the evidence that we have had. It is very powerful evidence that there are strong reasons, relating to promotion, getting resources and so on, to artificially report the crime figures—in fact, to report them as being lower than they are. How do you react to that evidence that we had in our previous session?
Jeff Farrar: It is interesting that you have heard evidence from a number of officers in the Metropolitan service—I know a chief officer from the Metropolitan Police who will be coming here—and some people who have not been in the service for some time. On that point about promotion for inflating figures, I can talk personally on this, and for me that would be a reason not to promote someone. I am running promotion boards in all ranks in my force at the moment, and if you asked, people would tell you that if you came forward with, “Look how I have reduced these numbers”, that would not be something that would be a feather in your cap. It would be the contrary, if they were not ethically done in the right way.
On what you have heard previously, I would say that I have talked to a lot of ACPO colleagues and others around the service since your first hearing. Unequivocally, people have been aghast at some of the things that were said. Take the comments on nodding, skewing and cuffing. Cuffing has generally been a term used in the service about somebody who is lazy and does not do their job; they cuff something. That is not managerially driven, but just not doing their job. On the other terms, every single chief officer I have bumped into has unanimously said that nodding was just completely unfamiliar to any colleagues. The description underneath it of nodding at houses could not happen now. You have to do a PACE-compliant interview; you could not write off crimes in that way. The system would not allow you to do it. On skewing, I think my colleagues would tell you that that is behaviour they do every day. They identify, based on the data they have, where crime trends are, and they deploy resources accordingly. The evidence I heard on that was that there is something wrong with that, but actually that is what every police force does every day. They move their resources around, based on demand within their particular area. Skewing, nodding and cuffing really are terms that are completely alien to me and to my colleagues.
Q272 Paul Flynn: I have just one question. Are you under more pressure now, with the regime of the Police and Crime Commissioner, to fulfil your work in order to get public approval by reducing those crimes that the public perceive to be the most important, rather than pursue your professional objectives of reducing those crimes that do the most damage?
Jeff Farrar: You have known me a long time; a lot of people have known me a long time. I have 30 years’ service, and if I thought I was under that undue pressure, I would not be in the service—absolutely not. Every conversation I have had with the Police and Crime Commissioner is about doing the right thing. That is the only service I could work in. I would leave the service if I felt that undue pressure and I was not able to do that.
Paul Flynn: I am grateful to you for that.
Q273 Chair: Your last HMIC audit of crime statistics was in 2011.
Jeff Farrar: Yes.
Chair: What proportion of those that they audited turned out to be misrecorded?
Jeff Farrar: I think it was 17.
Q274 Chair: Out of?
Jeff Farrar: I cannot remember the figure off the top of my head.
Chair: It is the proportion that matters, isn’t it?
Jeff Farrar: It was about 10%.
Q275 Chair: About 10%—which is exactly the evidence we heard from Police Constable James Patrick and the others. It is exactly in line with the expectation.
Jeff Farrar: But within those 17, there were lots of reasons why they may be misrecorded—
Q276 Chair: I understand that, but if 10% of crimes are not being recorded properly, is that not a cause for concern? It was in 2011.
Jeff Farrar: It is a cause for concern.
Q277 Chair: But, you see, some of what you are saying suggests you are perhaps too keen to justify what is good and to explain what is good, and perhaps not keen enough to accept that there is a problem, albeit in a minority of cases. To your credit, it seems to me you are making great endeavours to correct that in exactly the right kind of way. Would it not be better for all the chief officers to accept that the compelling evidence we heard last month was indeed compelling? It was backed by other independent witnesses, such as the Essex Police and Crime Commissioner, who said that even if we are trying to correct this, it is going to have a long tail on it. You are correcting a culture. How much would you say you are challenging and correcting an established culture with your code of ethics, with your good practice guide, and removing it? Is that not what you are doing?
Jeff Farrar: Absolutely. On the evidence you have heard previously, I know a lot of my Chief Constable colleagues were aghast at those circumstances, and they did not recognise that within their locality. It might be very difficult to assume, based on the evidence of some officers, whether there is veracity in that or not, that that is replicated across 43 forces, because we are all very different.
Q278 Chair: How unexpected is it, when confronted with an ugly truth, that Chief Constables will go into a huddle and explain to themselves that, “Well, actually, it’s not nearly as bad as that, is it, fellas?” How unexpected is that? Is that not what organisations do?
Jeff Farrar: I cannot comment for every Chief Constable in the country on that, but what I would say is my experience is that is not the way they would respond to something like this.
Q279 Chair: How confident do you think all Chief Constables can be that they are actually being told the truth, and that you are being told the truth? It is a very hard thing to accept bad stuff like this in the organisation.
Jeff Farrar: There is an inconsistency across the service in the message the chief officers send to their staff and to FCRs. In some forces, the Force Crime Registrar regularly meets with the Deputy Chief Constable or Chief Constable. In some forces, the Chief Constable or the Deputy Chief Constable might not even know their name or if they walk past them in the corridor.
Chair: Really?
Jeff Farrar: That would be the truth of it, yes, accepting that a force my size is a small force. I have 2,500 people. In a force that has 50,000 or 80,000 people, you would expect that to be different. That is a really important point about engagement.
Q280 Chair: We are running out of time here, but I asked you earlier what recommendations you would make about FCRs, and you said you had jotted down a few points. It would be very helpful for this Committee if ACPO would go away and think about this, and perhaps send us a more considered note about how FCRs should be treated by Chief Constables, the kinds of relationships they have with Chief Constables and how the FCR should be part of the leadership, so that the ethical leadership and the statistical leadership are as one. Is that not what we want?
Jeff Farrar: Absolutely.
Q281 Chair: That is not really what you have explained to me today. I do not think your policy is very well-developed.
Jeff Farrar: It would be very helpful if I shared the letter I wrote to Chief Constables probably about eight or nine months ago around Force Crime Registrars and their role.
Q282 Chair: What responses did you get to that letter?
Jeff Farrar: It varied from different forces. When I visited their Force Crime Registrars, a number of people thanked me and said that it had prompted it in their chief officers’ minds.
Q283 Chair: How fair or unfair is it to say there may be a cultural problem in the attitude to FCRs right at the top of some constabularies?
Jeff Farrar: It is what I started with: it is hugely inconsistent across the 43 forces.
Chair: You have been very helpful to us today, Chief Constable. We have asked you some pretty tough questions and you have been very honest with us. I am grateful for that. Thank you very much indeed. If you want to send us any further points for evidence, please do so.
Oral evidence: Crime Statistics, HC 760 2