To be published as HC 94-vi

House of COMMONS



Justice Committee


Tuesday 28 January 2014

Richard Garside, Ben Page, Professor Andromachi Tseloni and Professor Mike Hough

Evidence heard in Public Questions 323 - 379



This is an uncorrected transcript of evidence taken in public and reported to the House. The transcript has been placed on the internet on the authority of the Committee, and copies have been made available by the Vote Office for the use of Members and others.


Any public use of, or reference to, the contents should make clear that neither witnesses nor Members have had the opportunity to correct the record. The transcript is not yet an approved formal record of these proceedings.


Members who receive this for the purpose of correcting questions addressed by them to witnesses are asked to send corrections to the Committee Assistant.


Prospective witnesses may receive this in preparation for any written or oral evidence they may in due course give to the Committee.

Oral Evidence

Taken before the Justice Committee

on Tuesday 28 January 2014

Members present:

Sir Alan Beith (Chair)

Rehman Chishti

Jeremy Corbyn

Nick de Bois

Mr Elfyn Llwyd

John McDonnell

Yasmin Qureshi


Examination of Witnesses

Witnesses: Richard Garside, Director, Centre for Crime and Justice Studies, Ben Page, Chief Executive, Ipsos MORI, Professor Andromachi Tseloni, Professor of Criminology, Loughborough University, and Professor Mike Hough, Professor of Criminal Policy and Associate Director, Institute for Criminal Policy Research, Birkbeck, University of London, gave evidence.

Q323 Chair: Good morning, everyone, and welcome. We are very grateful to you for coming along to the Committee today and giving us the benefit of, respectively, your experience, your research or your polling in an area that we are studying carefully at the moment. We have Richard Garside, director of the Centre for Crime and Justice Studies; Professor Tseloni, from the University of Loughborough; Professor Hough, from Birkbeck; and the chief executive of Ipsos MORI, Mr Page. It might be helpful if you could go along the line and give a quick indication of your areas of experience and expertise before we start.

Richard Garside: I am Richard Garside. I am the director of the Centre for Crime and Justice Studies. The centre was originally established in 1931, so we go back quite a long way. We are an independent, non-party political educational charity. Our real focus is on informing the public and policy makers about the operations of the criminal justice system, the causes and consequences of, and solutions to, crime, and social harm. We very much ground our work in evidence, so we have strong links with the academic and research community. Finally, in addition to my role as director of the centre, I am professor at the Queensland University of Technology in Brisbane and senior research fellow at the Open University, so I wear a few hats.

Ben Page: I am Ben Page. I am the chief executive of Ipsos MORI. We are one of the largest research organisations in Britain. Among other things, we have looked extensively at public perceptions-or misperceptions-of crime and why that might be.

Professor Tseloni: I am Andromachi Tseloni, professor of quantitative criminology at Loughborough University. I am an econometrician by training, but for the last 24 years I have been working with large crime datasets, both national and cross-national. I have worked at universities in the States-the university of Maryland-Greece and here in England: Manchester, Hull and now Loughborough university. My research focuses on identifying risk and protective factors for criminal victimisation and fear of crime, and how these may interact or be conditioned by context or neighbourhood characteristics. More recently, for the last five years or so, I have worked with colleagues from UCL and Simon Fraser University to try to explain the recent crime drop.

Professor Hough: I am Professor Mike Hough from the institute for criminal policy research at Birkbeck. I started my career as a Home Office researcher and was on the team that designed the British crime survey, now the crime survey for England and Wales. Since then I have done quite a lot of research on policing and on attitudes to crime, sentencing and policing. Most recently I have done comparative work looking at trust in justice and perceptions of legitimacy of justice across Europe.

Chair: My colleagues may sometimes identify a particular witness they would like to answer, or it may be obvious to you that this is your own area, in which case please jump in and answer that question, rather than feeling that everybody has to speak on everything.

Q324 Mr Llwyd: Good morning. May I start with Professor Hough? Recently you have said that both the indicators-that is to say, the official crime survey and the police-recorded statistics-point in the same direction: that there has been a substantial decrease in crime over the last few years. However, you have since expressed an opinion that possibly the fall as recorded by the police is not quite as dramatic as they have been saying, for various reasons. There is now a debate about how much crime has fallen. I think we all agree that crime has fallen. The question is: by how much and to what extent? Do you have a view on that?

Professor Hough: Yes. I think the crime survey for England and Wales is probably a more reliable indicator over the last-

Q325 Chair: But you did design it.

Professor Hough: Because I designed it? No, but you must bear in mind that I am attached to the survey-although not in a work capacity any more. Nothing changed about its design over the noughties, from 2001 or 2002 onwards, and there is every reason to think that it is a stable indicator. There is every reason to think that the police-recorded crime system has changed, specifically in the proportion of crimes reported to the police that get recorded. ONS has given estimates of the shrinkage in the proportion of crimes recorded, which is quite marked. It was up at around 90%, but it is now 70%. It is fairly clear what the reasons for that are.

Q326 Mr Llwyd: Could you elaborate on that?

Professor Hough: Between 2003 and 2007-08 the Audit Commission audited police-recorded crime processes quite closely, and it was clear that the new national crime recording standard was being increasingly more accurately observed. The audit stopped in 2007. There were things like Sir Ronnie Flanagan’s report on bureaucracy and Jan Berry’s later report on bureaucracy, all of which were signals to the police that crime recording was not quite as important as it was in 2002-03. The consequence was that there was a lower recording rate of crimes reported to them. I think the evidence is fairly clear on that.

Q327 Mr Llwyd: I think we would all agree that there is a downward trend, which is good, but none the less there are some offences that-if I can put it this way-buck the trend; in other words, they have gone the other way. There has been an increase in some areas. Could you, or anybody else on the panel, describe what they are, and say why you believe this is?

Professor Hough: It is clear that e-crimes-e-enabled crimes, crimes of identity theft and so on-are growing, because that is a new area of criminality. There are some crimes that are probably better reported to the police now; sexual offences are the obvious example. There may be something in the argument that shoplifting is increasing now, according to the police statistics, because of simple need-but we need to look at the trends for longer before we can be certain of that.

Richard Garside: I agree almost entirely with what Mike has said, but I want to introduce a couple of caveats. First, if we look at the very long term, all sorts of crimes have fallen. Homicide in 14th-century England, for example, was about 20 times as high as it is now. There is a good news story, both in the long term and, as Mike indicated, in the shorter term. Over the last 20 years there have also been significant falls, burglary and car crime being really good cases in point. But in some other areas we have seen rises. Street crime, for example, is higher now than it was in the early 1980s-probably; we do not know that for sure. Although the public are often portrayed, at least in some circles, as somehow foolish for not understanding the good news about crime, the experience for some people will be that it has got worse-in the case of street crime, for example.

The distribution is also very lumpy. If you look at, for example, homicide between the early ’80s and the late ’90s, on average homicide rose, but in the richest areas your risk of being a victim of homicide fell. In the poorest areas, your risk of being a victim of homicide increased sixfold, so there is quite a significant difference in terms of both locality and different crime types.

One of the things that I often say is that, when we are talking about crime, talking about overall crime does not make a lot of sense. There is a lot of difference between shoplifting, on the one hand, and mass murder, on the other. Unless you think they all have the same underlying cause, it is important that you think about different crime types-how they may have changed and shifted over time, and what the underlying causes of the shifts might be.

Q328 Chair: Professor Tseloni, do you want to come in here?

Professor Tseloni: Coincidentally, we were in a meeting yesterday at the Royal Statistical Society that examined exactly the difference in the slope of the falling trend between recorded police crime and the crime survey data. While it was clear that for serious crime there is not any difference, there is, perhaps, a difference where the crimes are less serious. It could, therefore, be an effect of the targets policy of the police. The police officer would have the discretion to move on to the next call if they thought, for instance, that they were not needed there, and that the victim did not want to prosecute further. That is one reason. For serious crimes, the two datasets are not very different. This is also the case cross-nationally. If we look at the comparative cross-national data, when it comes to serious crime and crime survey data, the trends are pretty close.

With regard to the trends for different crime types, I agree with both speakers so far. We have to examine different crime types in isolation, not just the overall crime trend, because that is simply the average. For instance, theft of mobile phones increased up to 2004-05, when everything else was going down. There is a slight increasing trend again in the latest dataset. When we look at violence, domestic violence and acquaintance violence have been decreasing, but stranger violence has been pretty much stable since 1995.

Q329 Mr Llwyd: We know, for example, that fewer cars are stolen nowadays because of increased security-the cars are more difficult to steal than they were. What other categories or persons seem to benefit from the fall?

Professor Tseloni: With Professor Nick Tilley from UCL and Graham Farrell from Simon Fraser University, and the kind support of the Economic and Social Research Council, we have examined specifically the reasons for the fall in car theft, and we are now in the process of examining the reasons for the fall in burglary. For car theft, we have distinguished between theft of cars, where the main reasons were the introduction of in-built central locking and immobilisers, and theft from cars, where in-built alarms and central locking were more effective in thwarting these crimes. This is so cross-nationally, because there has been independent research done in the US, by Mike Maxfield, and in the Netherlands. We have also looked cross-nationally at the US and Australia. In all of these countries, car theft started dropping when in-built car crime security was introduced, allowing also for time to renew the car fleet in a country.

With regard to socio-economic groups, in analysis we did with Louise Grove from Loughborough University, we found that those who have benefited the most are people living in owned accommodation on higher incomes, not lone parents. I am sorry-can I look at my notes?

Q330 Chair: Of course. If afterwards you think of a point that you wish you had made on something like this, by all means write to us after the session.

Professor Tseloni: We examined exactly this question, which is important. People who own cars are another group who have benefited. You have to have a car to be more protected.

Chair: Yes.

Mr Llwyd: I think I follow that.

Professor Tseloni: What is important now is that people in the most vulnerable population groups are much worse off compared with others than before the crime drop. It seems that crime-or criminal victimisation-is concentrated on very few population groups. As a consequence, in my view, it is much easier to target and, if possible, to eliminate.

Richard Garside: I want to pick up on that point. If you look, for example, at the long-term violence trend according to the British crime survey and other crime surveys for England and Wales, roughly speaking from about the early to mid-1980s it went up, until about 1995, and then dropped back down again. However, if you break down the different violence categories you see a very striking picture. Andromachi mentioned earlier that stranger violence has not really changed. Across the whole period, stranger violence was just level-you were no more or less likely to be attacked by a stranger. Domestic violence, on the other hand, went right up-and back down again. It is worth bearing in mind that your gender, as well as where you live, has an impact on your risk of victimisation. Most victims of domestic violence are women, and most perpetrators of violence are men, so there is a very strong gender effect in relation to forms of violence. If you just look at violence overall, you will miss those very important distinctions.

Q331 Mr Llwyd: I do not know whether this is a fair question, but is it possible to map out geographically where the greatest fall has been? I dare say it is a very difficult question. I was told by a colleague who is a member of the Committee that in his constituency, in Islington, on one side of the road there is a high crime area but on the other side of the road there is not. Is it possible to give some kind of snapshot of where, across England and Wales, we are now seeing the greatest decrease?

Professor Hough: It should be possible to do that analysis. I do not think that the Home Office or the Office for National Statistics has done it, but Professor Tseloni may have.

Professor Tseloni: You can do it at the regional level or, possibly, at the level of police force areas, but with a very large error for police force areas. We have done it at the regional level.

Q332 Mr Llwyd: At the level of police force areas, obviously you would be looking at the police-recorded crime figures.

Professor Tseloni: No-it is with the crime survey for England and Wales data. Originally, when the sample was increased back in 2001-02, it was done so that it would be representative for each police force area. Now it is down again slightly, which would affect sampling error.

Q333 Chair: Don’t insurers collect this information by postcode?

John McDonnell: They published burglary rates for towns last week. I know, because my town came in the top 10, unfortunately. I think that was for insurance.

Ben Page: That will be helpful, but of course it will focus on people who are insured, so there will be some coverage issues.

Q334 Chair: But they make an analysis of where they think the risk levels are higher. That can be a bit arbitrary because, as in the example given earlier, the same postcode will include high crime and low crime areas.

Ben Page: In our work-looking at about 25,000 interviews last year-interestingly, it is older people, wealthier people, white people and Welsh people who are least anxious about crime at a whole. This is when we ask people whether it is a problem facing Britain, rather than a personal problem. Young working-class and, particularly, black and Asian people are most worried. Of course, in some ways young people are most likely to be victims of violent crime.

Q335 Chair: Has anyone studied the figures in an attempt to establish more clearly what displacement takes place from those crimes that become more difficult to do? We have had the example of car crime. Do we know whether there is significant displacement to another crime?

Professor Tseloni: We looked at car crime, and there was no significant displacement to older cars. Colleagues from the Netherlands have also produced a study, which I have submitted here, in which they looked at the effect of in-built burglary security devices in new homes from 1999, when those were mandatory. They did not find any displacement to older homes. What they found was that there was a 26% drop in burglary for the areas with newly built homes, compared with the older building stock, but this was diluted nationally. Over a period of 10 years, there was a 5% drop in burglary at national level, just because the housing stock was replaced. They did not find any significant displacement.

May I introduce here something that is close to my heart? The Secured by Design standards are now at risk of being removed from the planning and housing standards. Although it is really nice to have simple guidance, I wonder whether we should not also have mandatory standards for burglary security in newly built homes. Without those, we may see burglary go up in the future.

Q336 Rehman Chishti: On the issue of burglary, you said that it is a lot to do with housing and new build. In Medway in Kent, the area where I live, which has a population of 270,000, there are only two burglaries a day, according to Kent police. Does that mean that it is less to do with Kent police and more to do with housing development?

Professor Tseloni: Two a day, according to the police? I need to compare that. How large is the area? What does it mean in terms of risk?

Q337 Rehman Chishti: In terms of risk, it is 50 miles from London and has a population of 270,000, with high levels of deprivation, so it is not a leafy green area. Kent police say that there are only two burglaries a day in that area. Is that to do with policing or is it more to do with house building and the criteria you have just outlined?

Professor Tseloni: Because it is police-recorded crime, it would have to have some input from the police, because the police have to be notified and attend; that is the usual caveat. It would be interesting to see what kind of security there is for the houses that have been burgled and for the houses that are close to the burgled households.

Professor Hough: May I come back to the point about prevention and displacement? You can have displacement, but it is important to remember that you can also have diffusion of benefits. If you prevent one form of crime, you may have a spillover into other forms.

Q338 Chair: That is what I was thinking about; that is what I meant by displacement.

Professor Hough: No, I mean a spillover of benefits. If you remove a form of crime or reduce the opportunities for crime-say, for car crime-people may have the lower rungs of their career ladder removed, and they cannot climb further up the ladder.

Chair: I see what you mean.

Professor Hough: So yes, there may be displacement, but there may be benefits for other forms of crime when you squeeze down.

Q339 Chair: So you have undermined the career structure of crime.

Professor Hough: In a way, yes.

Professor Tseloni: This is a possible hypothesis to explain the drop in violence, which happened a few years after the drop in car crime and burglary. It could be due to the spillover effects from truncating offending careers at the car crime level. In the ’80s and early ’90s, it used to be very fashionable for young lads to go joyriding. If this is now impossible and opportunistic car theft is out of the question, they may turn their attention to something more legitimate, like playing sports.

Q340 Rehman Chishti: I want to move on from the category of "Is recorded crime really falling?" to "Why is recorded crime falling?" Can I get some explanations for this? To what extent can we now adequately explain why crime has reduced-first, in the United Kingdom, and secondly, internationally?

Professor Tseloni: It is important to note a few things. First, crime dropped internationally. This means that the drop cannot be explained by criminal justice changes within any specific jurisdiction. For instance, work by Professor Zimring found that crime fell in the US and Canada similarly, but in the US sentencing went up during the same period, whereas in Canada it stayed stable. Sentencing, therefore, is not an answer. It should be something that modern societies have introduced-changes that have happened in all those societies during this period. It should also be something that explains the previous crime rises-the hikes from the ’60s up to the mid-1990s.

It is also important to notice that in the US crime fell earlier than in Europe, so the explanation would be based on factors that appeared first in the US. One such explanation is opportunity. From the ’60s onwards, the economies were thriving and there were more consumer goods around, so there was more opportunity to steal things. This created a circle of fear of crime and then more target hardening, which appeared first in the US and subsequently in Europe. That is responsible for the crime falls we have seen since 1995 in Europe and since the 1980s in the US. It is explained in our statistical analysis of car theft and burglary.

It is also important to notice that, cross-nationally, car theft and burglary started falling first, at the beginning of the 1990s. Then, from 1995, it was theft from the person and theft from cars. Finally, from 2000 onwards, there was a fall in violence. This time line for the falls in different crime types could again be explained by the debut crime hypothesis, which Nick Tilley introduced, that removing the opportunity for car theft and burglary may truncate criminal careers. There is also less opportunity for violence related to the stolen goods market.

Ben Page: Obviously there is what the professionals around the table would call target hardening, but there is also the point-Mike will probably pick this up-about social change and acceptability. Societies have become more intolerant of violence, and more tolerant generally. I think you can see some of that. There are issues around the birth rate. There are even suggestions by some of my colleagues that mobile phones distract teenagers more than was the case when, say, I was growing up in the 1970s. It is a global phenomenon that is enormous in terms of its impact on society, but we still do not understand it as much as we might. There are things that you would notice, like better regulation of certain environments, but undoubtedly social change plays a part. Just look at attitudes to drink driving, homosexuality or anything else. There have been massive changes, generally in one direction.

Professor Hough: I think there are very long-run trends that reflect a trajectory towards civilisation; I am quite an optimist. There are specific factors driving particular forms of crime up and down. For example, the reduction in the consumption of alcohol in the UK from about 2005 is very probably a factor in explaining the reduction in violence. So you can look to crimes for specific explanations. These seem to interact with more general ones at the same time.

Richard Garside: I would urge the Committee not to look for a general theory of why crime as a whole is falling, because except at a very abstract level-which from a policy point of view is not particularly helpful-I suspect that you will not shed much light on the problem. I would break down the causes of crime falls-or, indeed, other crime changes-into three broad categories.

There are the superficial factors. Why is burglary falling? One of the reasons is that houses are more difficult to break into and cars are more difficult to steal. So there are some important but relatively superficial changes to the way that societies organise. There are some institutional questions. For example, a few years ago some interesting research highlighted an apparent link between the increase in the compulsory age of education from 15 to 16 in the early 1970s and a fall in property offences. That is about keeping young people in school. Raising the age to 18 now, and the general move towards more young people staying in education for longer, will-all other things being equal-probably have an impact on some forms of criminality.

That is not a crime reduction policy. It is really important that educational attainment and staying in school are not marketed as "Stay in school to stop being a criminal," for all sorts of very obvious reasons. Education is a good thing in its own right and should be considered as such. One of the things I am really interested in is Government policies that are not actually about crime reduction but none the less probably have a crime-reducing effect. The long-term fall in child poverty, for example, over the last 10 years will probably have an effect on levels of crime-or at least certainly on levels of some forms of offending.

That is the third thing-those big structural issues. Earlier I mentioned the changes in violence, particularly the impact on domestic violence through the ’80s and ’90s and into the noughties. Why is that? It is because we live in a society where far too many men think it is okay to hit women. At one level, it is as simple as that, but underlying it are some very profound questions about gender relations: what masculinity is in society, what it is to be a man in society and the relationships between men and women in society. Those are big structural questions that, clearly, are outwith the overall ambit of this Committee but are important to consider.

Q341 Rehman Chishti: Why has it always proved difficult to be clear about what causes reductions and rises in crime?

Richard Garside: I am not sure that it has, when you start talking about particular crime types. Last week you had a very clear exposition by Gloria Laycock about the reasons for falls in car crime, for example. If you try to answer the question, "Why is crime falling?" you are asking the wrong question. I suspect the Health Committee would not ask, "Are we healthier?" It might be interested in coronary heart disease, particular forms of cancers or particular mental health issues, but it would not ask about health in some kind of abstract, grand sense. Answering the question, "Why is crime falling?" is problematic when you ask the question, "Why is crime falling?" If you ask the question, "Why is burglary falling?" or, "What has happened to homicide rates?" you can start to shed some light on the problem.

Professor Hough: It is difficult to be certain. It is easy to speculate, but the problem is the counterfactual: what would have happened if that particular social change had not occurred? You can do comparative work looking across different countries, but it is very hard to nail it down, because of this difficulty in proving the counterfactual. That is how I look at it.

Q342 Rehman Chishti: So when you see an article that says, "Crime has gone down by 10%"-

Professor Hough: No, that is the measurement of crime. It is the explanation of the 10% fall that is the problem.

Professor Tseloni: With regard to car crime and burglary, the explanation is that there are now fewer opportunities. My research team is pretty clear about that; I do not know whether it is clear to everybody else. One reason that confirms that is that the population groups with less security than others are now much more at risk of burglary than they were before the crime drop in 1995.

Q343 Rehman Chishti: Moving on to a final question, if I may, why has the recession not caused an upturn in the crime statistics, as many criminologists had expected? Professor Hough, I refer you to a quote that you gave in 2013: "This fall is striking and unexpected, especially in view of the fiscal crisis, whose impact is bearing down sharpest on the poorest and most marginal social groups."

Professor Hough: I was one of the people who said that the recession would kick crime up at some stage. I have yet to be proved right. I suspect that different recessions have different cultural meanings, and that the ’70s and ’80s recessions meant different things to the people at risk of crime, compared with now, but I have not really got much further than that to explain it. I may be proved right, too.

Richard Garside: If you unpack the different crime types, the largest single crime type in the crime survey for England and Wales-Mike will correct me if I am wrong-is car crime. When you have such a large volume of that particular survey on a downward trajectory, which is probably nothing to do with the recession but is to do with changes in technology, even if other crime types were going up they would have to be big volume crimes to buck that overall trend. Some changes in crime types are probably countercyclical. They are not necessarily related to economic processes, at least not directly. They may be in the long, long term, but not directly.

This may be the case more for other things, such as violence. I will not be overly surprised if we see increases in levels of domestic violence, in particular, in the coming years, but that is not necessarily the case. Partly, it will depend on other factors, such as the degree to which cultural attitudes have changed and it is just considered to be less acceptable to hit women, and the degree to which women are more economically independent of men and so do not have to stay in an abusive relationship. Those kinds of things will cut across other factors.

As I said, if you try to answer the question, "Why has the recession not led to a rise in crime?" I think you are asking the wrong question. The question is, "What is the relationship between certain economic processes and certain types of crime?" You can then see some possible relationships-but, as Mike said, it is not at all clear.

Chair: Mr de Bois has a supplementary point.

Q344 Nick de Bois: Professor Garside, notwithstanding your warning not to look for generalities, are there any specific crime areas where the changing age demographic has meant that there are fewer younger people and more older people? Do you think that has an influence on figures? I may be wrong, but I am assuming that more younger people commit crime than older people.

Richard Garside: Certainly more younger people are drawn into the criminal justice system and processed as offenders. That does not necessarily mean that more young people are committing crime. There are people who regularly try to get my credit card details from some kind of dubious place in the world; I do not know how old they are. We have seen in relation to some issues around financial crime, corporate corruption and, indeed, state crime, that the people doing that kind of stuff are not necessarily 25-year-olds. For sure, those types of offences that may be related to younger people-

Q345 Nick de Bois: Car crime, for instance.

Richard Garside: Car crime, potentially-and forms of homicide. The largest age group for victims of homicide is the late teens and early 20s. Probably a lot of the time they are killing one another. It is worth bearing in mind that it is a relatively rare offence; none the less, changes in age profiles can have an impact. However, that will be offset by other policies. If you have a very inclusive education system, that will tend to offset some of the issues relating to youth crime, for example. Likewise, if you have a healthy jobs market, that will tend to offset some of the issues relating to youth offending and youth crime.

Ben Page: We have fewer young people, and they are actually better behaved than they have been for a very long time. They are drinking less and taking less drugs. Our young people today are just not as revolting as they were in the 1970s. That is almost certain. We may or may not be around to look at it in the 2020s-or we may be doing different jobs-but the population profile is changing. You have this cohort going through school now. With more young people again, will you see the figures go up? We do not know. We will wait and see.

Q346 Chair: I will come back to Mr de Bois shortly on another point. From what several of you have said in the last half an hour, one could draw the conclusion that there is not much point in the Government understanding better what the causes of crime are, because the range of factors affecting levels of crime is so diverse-unless you conclude that the Government should operate at a much more sophisticated level and look, crime by crime, at the wider policies that might influence crime. We have had some interesting examples. There are more obvious and well-known ones about the links between welfare availability and crime, but other, quite different, policies can influence crime. Is it pointless-or not-for Governments to try to study the causes of crime and to adjust policies accordingly?

Richard Garside: I think it is really very important-but it is also important, as you indicated, to be fairly sophisticated about it and to think about different crime types and what you want to do about them. I would broadly distinguish between criminal justice policies specifically-more or fewer police, more or fewer prison places, more prosecutions and so on-and broader crime reduction policies, including the work of the crime reduction partnerships, some of the stuff that local authorities are doing and things like the troubled families agenda, inasmuch as it is sometimes presented as being about tackling delinquency.

Those two can have some dampening effects in the short term, for sure, but their long-term impact is probably vastly overstated. There are then all sorts of other policies that Governments can engage in-educational policies, health policies and housing policies. Housing is a really significant issue, as we all know currently. Those are the kinds of areas in which, if Government are doing things that generally make society better and life more liveable, that will have the beneficial side effect of probably leading to reductions in some forms of victimisation. Again, pursuing those as crime-cutting strategies in their own right is probably the wrong approach, but the knock-on effect of Government policy can be quite significant over time.

Ben Page: If we believe the voters and the public-sometimes I do and sometimes I don’t, on this-every single time we ask them, they say that they believe that better parenting, rather than sentencing policy, policing or anything else, would do most to reduce crime. The one thing that everybody in Britain agrees on-we can find only one person in 100 who strongly disagrees-is that parents should take more responsibility for their children. Of course, they probably tend to mean that other parents should take more responsibility. Nevertheless it is striking that although the public love locking people up-they believe in sending people to jail and putting them on boats in the channel, if the jails are full-they do not believe that locking people up actually reduces crime. That is at the bottom of their list. They believe better parenting reduces crime.

Professor Hough: We have been talking as if the criminal justice system had no impact on trends. I think it does have an impact, but a smallish one. We should not forget the fact that you can buy reductions in crime through better policing, for example, and more imprisonment. The doubling of the prison population has been a very expensive way of achieving a small decrease in crime, but to say there has been no impact would be wrong. Similarly, I think small gains can be made by work with offenders. We should not bracket off the criminal justice system as something that shapes crime trends. The quality of the criminal justice system-how it relates to people-is probably as important as the quantity of criminal justice that it dispenses.

Q347 Chair: Internationally, there are countries with declining prison populations that also have declining crime rates, aren’t there?

Professor Hough: I am not saying that the prison population is a primary shaper of trends. I am saying that it is probably something you have to include in your model of what drives trends-ditto policing.

Professor Tseloni: We should not think of policies as being either education policies or health policies. In my view, better parenting is a health policy. It is in the interests of every society to raise a well-rounded new generation of citizens. If we look at this as a very narrowly focused crime reduction policy, perhaps we are missing a point. There are a lot of unintended crime reduction consequences of policies we would need to have as a modern, welfare-based system society.

Q348 Nick de Bois: Professor Tseloni, is the public’s fear of crime proportionate to the levels of crime? Could you summarise? In your opinion, what explains any disparity that I suspect exists?

Professor Tseloni: There has always been this idea that, on average, fear of crime is disproportionately higher than crime rates. Work by Brunton-Smith and Purgis from the University of Sussex-I did not do this work, so I have to reference the people who did it-has analysed fear of crime. They examined the exact relationship between different socio-economic groups and fear of crime, as well as how that is shaped by the neighbourhoods in which they live. According to their results-which I find totally trustworthy, following the analysis they have done-it seems that the public are very rational about fear of crime. Their analysis shows that, as crime rates in a neighbourhood rise, fear of crime rises. It rises more for single victims. If non-victims live in a no crime area or a very low crime area, they have no fear of crime. If they live in a high crime area, they have a certain level of fear of crime, but with single victims, as the crime rate in their neighbourhood increases, their fear of crime increases further. In the very high crime areas, it is double that of non-victims. The interaction between personal experience of crime and neighbourhood crime rates creates this level of fear of crime. For repeat victims, it does not really matter whether they live in a criminogenic area or in a non-crime-ridden area. They have a high fear of crime because they have experienced crime quite frequently.

Q349 Nick de Bois: To press the point, are you saying that whether a group of people are more fearful of crime is more dependent on whether crime has come to or exists in their area, or whether they live in an area where there is no crime? There is no other grouping you could think of-not younger, older or whatever it may be?

Professor Tseloni: Vulnerable people such as females and those with long-term illnesses are also more fearful of crime if they live in neighbourhoods with visible signs of incivility, such as low-level-

Q350 Nick de Bois: It is more accentuated.

Professor Tseloni: That is right.

Q351 Nick de Bois: Mr Page, do you want to come in?

Ben Page: We all know this, but it is the difference between the national and the local. Richard Garside is rightly pointing out that we need to look at specific sorts of crime if we are trying to know what will reduce them. If you really want to dig into crime, you need to start talking about people’s lived experience and what they think about locally, as well as the national debate about crime. The two are interconnected. People who read tabloid newspapers are much more concerned about crime than people who read broadsheets. Is that because the tabloid newspapers cover crime, or sensational crime, more, or is it because those people turn out to be more working class and live in more deprived communities, and might be more likely to be victims of crime? That is certainly there.

Q352 Chair: So there is a correlation; you have established that.

Ben Page: Oh yes. It is the same on immigration. The gap is bigger on immigration than it is on crime, but certainly the media have an impact. In the monthly survey that we do-the report we do every month, in which we just ask Britons, "What is the biggest problem facing this country?"-overall mentions of crime as an anxiety peaked in 2007, long after the actual peak in crime, with the murder of Garry Newlove and the shooting of Rhys Jones. Those were two signal crimes at the national level. We have not talked about signal crime yet, but it is hugely important. That leads to these sorts of gaps, so that 58% of people in Britain don’t think that crime is falling; they disagree with the figures. There are all sorts of reasons for that, and I guess there are all sorts of solutions that one could postulate.

Q353 Nick de Bois: Let me link the two points, if I may. Your organisation, Ipsos MORI, looked at the UK media disproportionately focusing on serious and violent crime. I think you demonstrated that 45% of crimes reported in newspapers involved sex or violence, compared with only 3% of crimes taking place. Are you saying that, apart from the localised groups Professor Tseloni talked about, we have a greater fear directly because of what we are reading in the newspapers? In your opinion, is that something that will drive policy?

Ben Page: It is certainly true that you cannot say for one second that there is a completely dispassionate reporting of crime by the news media. However, another report that we did looked at the views on a whole range of issues of people who never read newspapers and of those who read them, to look at what their overall levels of anxiety were. There we did not actually find much difference. Certainly media reporting of crime could not be described as balanced. Whether or not one believes in greater regulation-personally, I do not-there is a huge debate. There are so many issues on which the British public are way off-beam in terms of the facts of the country they live in.

Q354 Chair: Can you clarify what you have just said about the non-newspaper readers?

Ben Page: We looked at the people among the population who read newspapers-people who reported reading a daily newspaper. It is a report called "You are what you read?"; it is on our website and was written a few years ago. We looked at those who never read print media but may rely on TV news or radio news, which are slightly more regulated than print media, and at those who looked particularly at newspapers. We found that there was one issue where there was a big gap, but it was not crime.

Q355 Nick de Bois: I am conscious of time, so may I move on to Professor Hough? I do not tweet Government statistics very often, but recently I did tweet the British crime survey statistics, which turned out to be a bit of a mistake. I was reined in by people telling me how disconnected I was from reality, saying that these statistics were absolute rubbish and so on; we have touched on this. In your opinion, to what extent is the perception among members of the public that crime is not falling related to a mistrust of statistics? Are we beyond saving? Will anyone believe statistics? What is your experience, and what are your thoughts, on that?

Professor Hough: I suspect that people are pretty suspicious of Government statistics across the piece, and that this is part of that general problem. The media focus on good stories-and good stories are about crime rising. They do not focus-at least, not until very recently-on crime falling. I suspect that is the driver of this extraordinary statistic that most people think crime is rising when it is not-except in their area. In their area, people have a sense that crime is going down.

Q356 Nick de Bois: But it seems that that is a phenomenon we are going to have to live with. Mr Page, it will not go away even if we say that no Government spokesman should ever comment on figures and they should all be independent. Is it too late?

Ben Page: I think it is incredibly hard. You have 7% of the British public who believe that politicians use official statistics accurately. Asked whether they have confidence in their own ability to understand statistics, 36% admit that they do not, and we suspect that that is an understatement. However, I think there are things that can be done. One habit of politicians of all parties is to focus on what we would call vanishingly rare statistics. An example from another area might be the very small number of households claiming over £100,000 a year in housing benefit. I think the figure was five. Of course, that gets reported. It is five, but it is five out of a population of 63 million. Of course, it helps a politician make the arguments, but you could not really argue that it is balanced. It is the same with murder. Humanity is naturally drawn to bad news-literally-and we go and deal with it, but I think there are things to be done about trying to call for a more balanced argument.

You might argue that pensions is a less exciting topic than crime, and the parties have managed to talk about that without some of the "he said, she said" type of activity that makes things worse, because you end up with reputable sources such as the British crime survey just being seen as yet another statistic. However, we have found that it is incredibly difficult to shift people’s views. If you show them the actual numbers and tell them that they come from an independent charity, they do believe them a little bit more than if they are from the Government, which may or may not be the right thing to do; official statistics are very carefully compiled.

I think organisations such as Full Fact and the UK Statistics Authority should be encouraged to try at least to create a level playing field and some basic facts we can all agree on. What I would find unpalatable is our becoming more like America. There was a great piece in the Chicago Tribune during the 2012 presidential election that talked about the death of Facts, along the lines of "He was ill for a very long time, but has now finally left us". There are some things that we can agree on.

Richard Garside: The point about the way that politicians use statistics is important. One of the good things that happened to crime statistics some years ago is that they were all published in one block. There was a time when you would have the British crime survey published, with crime going down-great. Two weeks later the police crime data would come out, with crime going up. Unsurprisingly, members of the public-never mind parliamentarians-were thinking, "What on earth is going on?"

The Home Office did some very good work in the late ’90s and into the noughties to try to pull those datasets together and explain their differences. You had the chief statistician regularly meeting journalists, talking through the figures and explaining the differences. I like to think that the current reporting on crime statistics, which is generally probably better than it was-certainly a decade ago-is partly the result of that shift in the way that the statistics were compiled, communicated and explained. The police data are now slightly up in the air because of questions over those, but the long-term direction of travel has been very positive.

Q357 Nick de Bois: Professor Hough, I have one last question for you. Can you give me your opinion on the extent to which public trust in prisons, probation and the courts is measured or monitored? Do we do that?

Professor Hough: The crime survey for England and Wales has a trend-quite a short one-on trust in the criminal justice system. The broad picture looks pretty stable, with some indication of trust in prisons, trust in courts and trust in probation increasing. The long run trend that we have is on confidence in the police, which shows a very long run fall until the mid-noughties, followed by a recovery. The important thing to bear in mind is that probably beliefs about and trust in the police flow through into trust in the rest of the criminal justice system. We have done an analysis of what the drivers of trust in courts and trust in probation are. People’s views of the police are a significant factor. So the police-

Q358 Nick de Bois: -are really seen as the criminal justice system in the eyes of the public.

Professor Hough: And they have an effect that washes through the rest of the system.

Q359 John McDonnell: Can I pursue the trust issue? Our whole debate seems to be about understanding and trust, both by ourselves and by the general public. From the papers we had, I thought there had been a fall in trust until about 2003, when it stabilised. Is that true? What are the reasons for that?

Professor Hough: I do not think we have consistent measurement of trust in courts and trust in prison and probation, but I may be wrong.

Ben Page: We have asked about it since 1983. I looked up the figures on the way here. In 1983, 77% trusted the judges and 61% trusted police to tell the truth; 18% trusted politicians to tell the truth. In 2013, 82% trusted judges, 65% trusted the police and 18% trusted politicians. So it is pretty flat for all those groups.

Q360 John McDonnell: In terms of the priority of the issue with the general public, since your "Closing The Gaps" report of 2008 has it remained a top priority?

Ben Page: No, it has gone down. This is partly related to concerns about the economy, but very recently we have had some of the lowest levels of anxiety-at least, spontaneous anxiety-about crime at the national level, as opposed to the local level, that we have ever seen. At one point in the last few months it fell to 15%, so it really is not there. Is that because of media coverage? Is it because of genuine falls in crime? It is pretty much at an historic low in terms of levels of anxiety, compared to that peak, as I said, just six or seven years ago, in 2007.

Q361 John McDonnell: Going back to some of the issues that you raised before, for the record, could you say what we know about what the public think is effective in reducing crime, as against what the research says is effective? Where are the disparities?

Professor Hough: People want to see more police on the beat, and tougher sentences. Those are the most salient things on their wish list. I think there is a criminological near-consensus that tougher sentences will not hit the spot. It is more contestable whether things like neighbourhood policing will have an impact. The public’s model of policing is a fairly straightforward, simple deterrent one. The reality is that probably people’s behaviour is shaped by the legitimacy of the justice system, the broader legitimacy of the political system and other factors to do with parenting and schooling, but people think in deterrence terms.

Q362 John McDonnell: Yet parenting comes out quite high.

Ben Page: When you offer it as a choice.

Q363 John McDonnell: So when you diagnose the problem.

Ben Page: If you add things like better parenting to the list, it comes top, but if you ask people, "Do you think sentences are too tough, about right or too lenient?" they will say that they should be tougher. When you expose them to what the actual sentences are for particular crimes-once they look at it-their views sometimes change; my colleagues will probably talk about that. In the broader sense people do talk about society as a whole, and it is not just about the criminal justice system.

One of the interesting things I find in public opinion is that they like strong sentences and locking people up, but they genuinely don’t seem to think it actually reduces crime. Obviously it is the punishment part of it that they like. In answer to questions about cuts to the criminal justice system after 2010, the most popular area of cuts is reducing education in prison, when that is offered on a list. The least popular, of course, is reducing the number of police.

Q364 John McDonnell: Does the visibility of those cuts have an impact on confidence in the system?

Ben Page: It hasn’t so far. The cut that people have most noticed so far since 2010, at the aggregate level, is potholes. Among users of specific services, it is care for the elderly.

Professor Tseloni: The international crime victims survey, which is the only comparative crime survey that exists-since 1989-has been questioning them on a sentence that ought to be given to a hypothetical young lad who stole a TV. The answers that the public gave cross-nationally were less punitive than the actual sentence that would be given for that crime.

Professor Hough: There is a particular paradox here. People think sentences ought to be toughened up, and actually they sentence rather like magistrates. They are quite surprised when they hear Crown court sentences.

Q365 John McDonnell: That is interesting. So the general public view, which is what you are trying to educate us about, is a very deterministic view, from a general analysis, but when there is a debate around specifics it can be quite a sophisticated response.

Professor Hough: Yes, precisely.

Richard Garside: Some years ago we did some work with the Magistrates’ Association; it was called Local Crime: Community Sentence. It was work presenting scenarios to the public, a bit like what Andromachi has just described, and asking, "What kind of sentence would you give?" It then went through some of the mitigating factors-maybe he was only a young person, with a difficult family background, or whatever-and the process of how decisions are made. Fairly consistently, once individuals had been taken through that process, they would often revise down their initial assessment of what an appropriate sentence would be. So the role of education and of public engagement in institutions is really important. Obviously, it is really important that the public have trust in public institutions, given that the criminal justice process as a whole has a relatively small impact on underlying crime trends, although I agree with Mike that it is not completely insignificant. There is an argument for saying the public have too much trust in the criminal justice system as an answer to problems, so one needs to balance that out-but clearly public trust in public institutions is important. Which institution is delivering what, and to what end, is an important qualification.

Q366 John McDonnell: I understand the argument around the trajectory towards civilisation-we all support that, I hope-but the argument that you are putting is that a more thorough understanding elicits quite a sophisticated response. The importance of politicians not using statistics in a way that undermines the ability to comprehend the system is therefore quite critical.

Professor Hough: We think in terms of people having top-of-the-head attitudes that they often produce in response to surveys, but if you scratch the surface you get more considered views. People we talk to say, "Yes, these are really complicated issues." They are not simpletons, but they have a top-of-the-head response to quick questions.

Chair: A couple of my colleagues were greatly delayed, I understand, on the Victoria line this morning, which meant that they missed the earlier part of the questioning. I will give them the opportunity to ask a couple of supplementary points.

Q367 Jeremy Corbyn: Thank you, Chair. Yasmin and I had a very interesting discussion about justice matters at Green Park station. Had you all come to Green Park station, we would have been happy to question you there; it would at least have been different.

You have all studied very well issues of perceptions of crime. I have two questions. First, do you try to engage with the media about the reality of crime statistics as opposed to perceptions, because we have uniquely bad media in this country, which are obsessed with violence and sex to the exclusion of almost anything else? Secondly, do you see any big difference between recorded crime and what I think is very interesting-the reality of crime that the British crime survey shows up? There are often quite big discrepancies between those two areas of statistics.

Richard Garside: As an organisation we do a lot of that work. Indeed, when the latest dataset came out I spent a number of hours on the phone to various journalists talking about the statistics-what they mean and do not mean. I am sure Mike did much the same; he does a lot of that work as well. It is really important to do that work, and for journalists to trust that what you are doing is explaining the statistics rather than trying to spin them a line or sell them a particular agenda. We take that work very seriously.

As I mentioned earlier, putting all the statistical sets in one place and then trying to explain the differences has been beneficial. In relation to the difference between recorded crime data-the police data-and the survey data, one way of looking at it is that those data are compiled in different ways and tell you different things. I often say that trying to make sense of crime in your area according to the police statistics is a bit like going to your local accident and emergency department and seeing who turns up. You would then think that you have a major problem in your area of kids with saucepans stuck on their heads and drunken people who have got into fights. The police data tell you what incidents the public have gone to the police about and reported to them-or sometimes, what incidents the police have come across-and, in addition, that the police have recorded and, in addition, that they have recorded in such a way that it conforms to particular requirements from the Home Office in terms of the notifiable crime stats.

What you have in the crime survey is people being approached in their homes by researchers and asked to recall incidents that happened. There are issues there as well. A significant number of the population say, "I can’t remember whether anything happened," so the survey data are not as much of a gold standard as Ministers sometimes imply.

Ben Page: I think the trends are clear.

Richard Garside: Yes, the trends are clear. For reasons I have gone into earlier, there are different trends. One of the problems is that there has been a tendency to say, "If we take the police data and the survey data and they are both going in the same direction, that means crime is falling. If they are both going in a different direction, that means crime is rising. If one is going one way and one is going the other, we need to try to explain the divergences." Actually, they are just different datasets, compiled in different ways for different purposes. There is a degree of comparing apples with pears going on when people try to balance one off against the other.

Professor Hough: On the press, it quite easy to have continuing dialogue with the broadsheet media and the broadcast media, but I very rarely get approached by anybody else. The red tops are not really interested.

Q368 Jeremy Corbyn: Do you approach them?

Professor Hough: I don’t approach; I wait to be talked to. That could be a problem.

Jeremy Corbyn: You might have a long wait, with some of the press.

Q369 Chair: They are studying the criminal justice system more closely.

Professor Tseloni: Universities have press offices now. If we come up with some interesting findings, they will issue a press release and send it around. It is then up to the newspaper to do a more in-depth interview or to talk to the researcher.

Professor Hough: But it is the broadsheet press that get interested in our sorts of stories. That is the sad reality.

Q370 Jeremy Corbyn: I think that a more proactive approach is a good idea. I meet my local police-as I am sure colleagues do-every month or so. We discuss their policing objectives and what their priorities are, and they are often wildly at variance with what the local and regional papers are reporting at any one time. I end up in the position of defending the police on their policing objectives, because they are looking at the objective issue of crime in the area rather than at whatever the local papers report. I wonder whether there should not be a much more proactive approach by those, such as yourselves, who seriously look into issues of crime and public perceptions of it.

Ben Page: I think a vigorous, and perhaps, better-funded, organisation to try to speak truth unto power, without fear or favour, to the media and politicians on their use and reporting of statistics would be a good thing. There are various organisations doing that, but we think that would be desirable.

Leaving aside what is in a newspaper, it is true that very specific local information about actual incidents of crime and what is being done about it almost at a neighbourhood or street level is beneficial in the work that we have done. I used to live on a square where the police would say, "There have been no burglaries this year, there was one last year, and the year before there were two." It just gives you a sense of being able to make your own judgment about where you live. Of course, the resources to do that need to be considered; some of the initiatives on crime mapping may or may not help with that. I do think that an independent body that was cross-party and tried to talk about flagrant misuse of statistics in this area, particularly with the red tops-or at least tried to get the red tops to pay attention-would be beneficial, just to shame them.

Q371 Jeremy Corbyn: An office of statistical responsibility?

Ben Page: Yes. I would say that; I like statistics. Perhaps we will see a gradual education in this area.

Professor Hough: As Ben said, bodies such as Full Fact are doing a really good job.

Q372 Yasmin Qureshi: Correct me if I am wrong, but irrespective of what bodies are doing the work, isn’t the real problem here that, when you do approach the tabloid press to give them the facts and figures, they are just not interested?

Ben Page: To be quite honest, if a tabloid rings me I turn on a tape recorder. My limited experience with the tabloid press has been that they are not that interested in facts.

Q373 Yasmin Qureshi: That is exactly what I was going to say-so it does not really matter how accurate the statistics are. Most people read the tabloid papers on a daily basis. Quite often they print a load of lies and misinformation; we know that. Frankly, they do not care, as long as they get a picture of a naked woman, some threesomes or whatever. That is what they are interested in, isn’t it?

Richard Garside: In defence of the tabloids, we do not do as much work with the tabloid press as we do with the broadsheets and the broadcast media-for sure. As Mike said, they are generally not as receptive. However, we have done some good work with the tabloid media. I would not automatically assume that they are not interested in the facts or in presenting them to their readership. To a degree, perhaps, they play a party political game in a way that the broadsheets do not. The tabloids that are strongly aligned with the Labour party, for example, are always looking for good knocking copy in relation to the current coalition Government. Likewise, when Labour was in power it was the same with those that are more aligned to the Conservatives. That does have an impact. We have certainly written comment pieces for the tabloid press on crime stats, and they have not asked us just to write them in a way that they would want to. However, those opportunities are not as great as within the broadsheet media; that is for sure.

The other thing, of course, is the online stuff. So much more content now is online. Broadly speaking, notwithstanding the fact that the Daily Mail is possibly the most viewed website in the world-or something extraordinary like that-in my experience tabloid media web presence is just not as developed and sophisticated, in some ways, as the broadsheet media can be, so there are fewer opportunities.

Q374 Yasmin Qureshi: I know that the average constituent of mine will be reading The Sun, the Daily Mail and the Express. A lot of them may have the internet, but they will not really go on it for that. Some will, but the impact that the print media have is far more significant. The internet is there, but on a daily basis it is still the print media that most people look at first thing in the morning, over their cornflakes.

Chair: Is that a question to Mr Page about his earlier survey comparing readers with non-readers?

Ben Page: We found that at the aggregate level there isn’t a difference, although the British crime survey says that people who read tabloids are more worried about crime than people who read broadsheets. Is that a function of their reading a newspaper, or a function of their profile and who they are-being poorer, more working class and, therefore, probably more likely to live in areas with high levels of crime?

Undoubtedly people are swayed by media coverage; I give you the coverage in 2007, which, as I said, was the peak in anxiety about crime. In defence of the tabloids-as I have said that I sometimes find them difficult to deal with-mostly they are not making up the child murders and everything else that they report, but this is about the relative amount of attention given to those. We are back to these vanishingly rare statistics and the focus on those relative to what is happening overall. I think that sort of balance, and how one tries to achieve it, is particularly important.

Q375 Chair: Can I put it to you that politics is not the only issue that drives in this direction? It is also that indignation appears to be a selling point.

Ben Page: Yes.

Q376 Chair: Having a headline that makes people angry because they think that somebody has an inadequate sentence seems in their minds, at any rate, to be the kind of thing that will cause readers to continue to buy their newspaper.

Professor Hough: Yes, I totally agree. News values are not driven by politics alone. The politics impacts on them.

Q377 Chair: All this, in turn, can have an influence on the political system as well. When this Committee produced its previous report on justice reinvestment, we were seeking to create a space in which we could have a more evidence-based discussion of what works in the criminal justice system. Do you think that that has happened, and that the political rhetoric has changed significantly?

Professor Hough: I do not see it at the moment. If people get used to the idea that crime is not out of control, it will put less pressure on politicians and may create the space you want, so it may be possible for the future. I also think that centres that are set up to marshal evidence, such as NICE and the What Works centre for justice, may prove significant, but at the moment I do not see a big shift.

Q378 Chair: Would anyone else care to comment?

Richard Garside: Again, I would make a distinction between criminal justice policies and other areas. In relation to prisons and policing, certainly, there is not an awful lot of evidence of evidence-based policy and practice in those institutions. There is to a degree in probation, possibly. Probation has applied quite a lot of the lessons of research evidence over a number of years. However-partly because they are so subject to point scoring and other things-I think those kinds of explicit criminal justice policies have been a bit tricky.

In some other areas-you can go right back to the Crime and Disorder Act 1998 and the impetus for local authorities to play a more active role, working in partnership, or you can think, more recently, about things like the MAPPA arrangements with probation and a broader collection of local partners, seeking to manage certain identified problems-you could argue that there has been some sign of progress. For sure, if you read the latest journal article on the relationship between homicide and underlying social processes, there is not much evidence of that feeding through into the policy arena, but that is probably partly because the journal article is written in such an inscrutable way that you need a PhD in statistics even to understand what the article writer is saying.

Ben Page: I would just observe that the lower level of overall public anxiety about crime at the aggregate, and almost, I would say, political, national level at the moment-because it is all about immigration, the economy and the cost of living crisis in various ways, if you are look at our monthly poll-in a sense gives both, or all three, parties a space, if they wish, to have a less passionate and more dispassionate argument about some of this. As we now run up to elections of various kinds, let us see if we can maintain that.

Q379 Chair: On a more technical point, the chairman of the UK Statistics Authority has observed that it is likely that recorded crime will increase as a result of efforts to improve the legitimacy of crime statistics. If that happens, does it not have all sorts of implications for other policies, such as payment by results, depending on whether something is classed as an offence and recorded as such?

Professor Hough: Yes. It will clearly create problems for any metric that rewards performance according to stats. I imagine that HM inspectorate of constabulary will launch a programme of audits to improve the quality of crime recording. That will certainly drive the recording rate up and may turn it into an upward trend, even if the British crime survey continues down. Where that leaves things like payment by results is very hard to see.

Richard Garside: Ultimately, it depends on how the payment-by-results arrangements are agreed. That is still very much an open question. You can make a distinction, for example, between crime that is recorded and convictions that are secured. If payment by results is more focused on reconviction rates, that will not necessarily have an impact in relation to recorded crime rates; they are two separate datasets. In the long term it might, because, if recording practices are going up, that may lead to more people coming to the attention of the police and more prosecutions. However, my understanding of some of the problems with the police crime data has been not so much that information has not become available to the police but that they have just not been recording it, so they will continue to know who is around and what allegations have been made.

I am strongly in favour, in one sense, of an increase in the amount of recorded crime, and of the police being encouraged to report and to record in a consistent manner offences that come to their attention-and in a fair manner, as well, so that you do not have them recording an incident as something when it is really something else because that helps with their internal targets. I would be more concerned about the impact in terms of policing and policing targets than about the payment-by-results stuff. I think payment by results is bonkers in other ways, but in terms of the police-recorded data there will not necessarily be an impact in the way that some fear.

Chair: Perhaps I should reveal, in conclusion, that about 20 years ago I was summoned to appear as a witness in a police disciplinary inquiry when a police officer was accused of having revealed information about the way crimes were being recorded. It included the disappearance of a rabbit hutch, with rabbits in it, which was, in the police terminology, "recorded off," because the rabbits could have escaped. How they would have removed the hutch was not made clear. In other words, this is an old problem.

John McDonnell: Years ago there also used to be the problem of offences taken into consideration. Prisoners would be visited in prison, and that is the way you would clear up a number of crimes.

Chair: Thank you very much, all four of you, for your help this morning. We are very grateful for it.

Ben Page: May I put on record my thanks to the Clerks to the Committee, who have been seminally well prepared compared with other Select Committees that I have sat in front of?

Prepared 3rd February 2014