UNCORRECTED TRANSCRIPT OF ORAL EVIDENCE
To be published as HC 94- v

House of COMMONS

Oral EVIDENCE

TAKEN BEFORE the

Justice Committee

Crime reduction policies: a co-ordinated approach?

Tuesday 21 January 2014

Professor Cynthia McDougall, Professor Gloria Laycock, Professor DavId Farrington and Professor StePhen Farrall

Evidence heard in Public Questions 271 - 322

USE OF THE TRANSCRIPT

1.

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.

2.

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.

3.

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

4.

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 21 January 2014

Members present:

Sir Alan Beith (Chair)

Steve Brine

Jeremy Corbyn

Christopher Chope

Mr Elfyn Llwyd

Andy McDonald

John McDonnell

Yasmin Qureshi

________________

Examination of Witnesses

Witnesses: Professor Cynthia McDougall, Professor of Forensic Psychology, University of York, Professor Gloria Laycock, Department of Security and Crime Science, University College London, Professor David Farrington, Director of Research and Emeritus Professor of Psychological Criminology, Institute of Criminology, Cambridge University, and Professor Stephen Farrall, Professor of Criminology, University of Sheffield, gave evidence.

Q271 Chair: Good morning and welcome Professor McDougall from York, Professor Laycock from UCL, Professor Farrington from Cambridge and Professor Farrall from Sheffield. Thank you very much for agreeing to come and give evidence to us today to help us in our inquiry, which I always like to think of as a return to the work we did in our Justice Reinvestment inquiry in the previous Parliament, because, in essence, we are looking to see to what extent the policies have been pursued that were set out there and, if they have, whether they were right or whether they have worked, and what policy developments we should now be looking at. It would be helpful, I think, if I invited each of you very briefly to indicate what your own particular field of research and experience is. May I do that? Let us start at this end of the table.

Professor Farrington: Starting with me, I am Emeritus Professor of Psychological Criminology at Cambridge University, and I have been there for more than 40 years. My research has been mainly on identifying the development of delinquency and important risk factors for delinquency, childhood and adolescent risk factors, and the extent to which they can be targeted in programmes to reduce crime. So I am interested in early intervention. I wrote a book called "Saving Children from a Life of Crime," which set all this out in gory detail. I have also been involved in the Campbell Collaboration, which tries to do systematic reviews of the effectiveness of interventions, cost-benefit analyses and various other things. I believe very much in a scientific criminology. This is one of the questions we were asked. I think we should have observation and experiment. We should have replication and valid and reliable measurement, and we should have randomised experiments to control for selection effects.

Chair: Thank you very much.

Professor McDougall: I am Cynthia McDougall. I am a Professor of Forensic Psychology at the University of York, but I have academic and practitioner experience. I started my career as a probation officer working in the community. I have worked as a forensic psychologist in prisons. I have worked with the civil service as Head of Psychology for Prison and Probation Services. I have worked at the university, setting up a Centre for Criminal Justice, Economics and Psychology. I set up a forensic psychology course at the University of York for training forensic psychologists. I have done quite a lot of research for the Ministry of Justice. I have done consultancy for probation trusts and for the Prison Service. I am committed, also, to rigorous research methods. I am very keen on those, and I am very keen on integrating those with practice. I think that is an essential feature.

Professor Laycock: I am Gloria Laycock. I am Professor of Crime Science at University College London. I spent 10 years in Wormwood Scrubs-as a psychologist-

Chair: I am glad you amplified that sentence.

Professor Laycock: -and then about 30 years in the Home Office, leading as Head of the Police Research Group. I then went, eventually, to UCL via the United States and Australia and set up the Jill Dando Institute of Crime Science. One of the things I am now doing is directing the consortium of eight universities to support the What Works Centre for Crime Reduction, which is based at the College of Policing. My interests are in crime reduction, crime prevention and particularly policing. I am very keen, as crime science implies, on experimentation and particularly on theorising about how something works-not simply "does it work?" but how and in what context.

Chair: And Professor Farrall.

Professor Farrall: I am Stephen Farrall. I am a sociologist by training. I am Professor of Criminology at the School of Law, Sheffield University. My main areas of research have been on why people stop offending. Like David, I run a cohort study-nowhere near as long or as impressive or large as David’s-of people who were supervised to probation in the late 1990s, following them up. We have recently finished a fifth round of interviews with these men and women and have just about put the finishing touches to a third book in that series on their experiences, charting their progress towards desistance, that is to say away from involvement in crime, and the role in which probation supervision does or does not play a part in those narratives. As well as doing that, which is very much a longitudinal qualitative research strand, I also have started over the last five or six years to explore the longterm impacts of the policies from the 1980s that you might think of as being Thatcherite and the impacts that they had on social economic experiences in the UK, and the impact that that will have had on people’s experiences of crime, both as victims and offenders.

Chair: What I am going to ask my colleagues to do when we start the questioning is this. Except where they indicate that there is a particular person they would like to answer, I hope that you will selfselect, and whoever feels they have something to say about it will contribute. Don’t feel everyone has to talk about everything, but, if you want to add something to what someone else has said, just make it clear to me that you want to do so and we will do it relatively informally so that we can get the best value from the experience you can bring to us. I am going to ask Mr Brine to begin.

Q272 Steve Brine: Thank you very much. Those instructions are very useful, actually, because I probably will direct these questions at Professor Farrington and Professor Farrall, based on what you have just said, but do indicate, ladies, if you have something in particular you would like to say.

Crime is reportedly falling. We did a big report on youth justice last year in which we said this. To be honest, if you read the report, we were saying in there that we can’t be absolutely sure at this stage why the number of entrants into the youth justice system is falling. We have a lot of witnesses coming in next week and we will be discussing this in more detail then as well, but in your view which aspects of current Government policy do you think are responsible for this fall or does it predate current Government policy?

Professor Farrington: If you look at the police figures and the victim survey figures, the decline appeared to start in about 1995 as far as we can see, which coincided with declines in many other countries at the same time. However, the repeated selfreported delinquency surveys do not show a decline. I do not know if you have looked at the report called "A Fresh Start to Tackling Youth Crime," which came out recently. That reviews a number of repeated selfreported delinquency surveys with young people aged 10 to 17, and shows no decline in offending according to their reports, so I throw that into the mix.

In terms of why there is a decline in crime, it is clearly something that is affecting many different countries because we see the same phenomena in many European countries, and in north America as well, so it has got to be something fairly general. One factor, clearly, would be increasing security. There is no doubt it is harder now to commit property crimes than it was. It is hard to say how far the increasing programmes have had a desirable effect. They may have. We have had increasing numbers of programmes targeting young people and they may have had a desirable effect, but the answer is that it is very difficult to know why crime is falling.

Q273 Steve Brine: Professor Farrall, do you have thoughts?

Professor Farrall: The question "Why is crime falling?" is the second of the two questions which I would ask. The first would be, "Why did it go up in the first place?" There is an assumption that crime has always been high. In fact, it hasn’t. If you look at crime trends, reported crime, which goes back in England and Wales to the 1860s, pretty much plateaus at a fairly low level until you get to about the 1960s. It then starts to rise gently through the ’60s and ’70s. You get to the early and mid-’80s and you see a real takeoff point, rising, as David said a minute ago, to the mid-1990s. Then we see the now familiar decline. So you have, if you like, something that looks a bit like a camel’s back. It is a hump. The question "Why has it gone down?" needs to be preceded by the question "Why did it go up in the first place?"

The modelling which we have done, which takes data from the 1970s through to 2006, suggests that things like levels of unemployment or rates of unemployment, levels of economic inequality, which is the standard measure we use, the Gini coefficient, and social welfare benefits are all associated with property crime. The first of those-more unemployment and more economic inequality-drives up property crime, which is kind of understandable. Increases in welfare spending drive it down, so I think the question as to why crime is down needs to come after the question of why it went up.

I suspect that the reasons that it may have gone up in England and Wales at the time when it did go up might be something to do with sudden economic restructuring. Economic historians trace processes of de-industrialisation back to the 1960s, and the 1950s in some instances-and certainly I wouldn’t argue against that. The issue is whether that suddenly accelerated in the early 1980s. Economic thinking suggests that it did, particularly between ’80 and ’81. That really kicked away the last remaining props of the industrial sector, particularly in the midlands and north, producing high levels of unemployment, which led to high levels of economic inequality. I suspect that some of that lump, of which we are now all thinking about the downside, in some respects is good. We are focusing on the positive-"Why is it coming down?" The alternative is, "Why did it go up?" It may be-it may be-about economic restructuring.

Q274 Chair: Does geographical distribution support that?

Professor Farrall: We are just about to start doing that. The problem with the research that we have done thus far is that it is at national level, that is to say England and Wales. What we are doing now, with money from the Economic and Social Research Council, is using selfreport data from things like the British Crime Survey, which is regionally coded, and the British Social Attitudes survey, the Labour Force Survey and the General Household Survey to look at those trends as near as one can do at the Government regions of the offices. So in two years’ time I will be able to give you something of an answer, but right now I can only tell you that we are going to do that.

Q275 Steve Brine: You can come back. Just going back to the point you said about welfare spend, you are suggesting then that when the Government hand out money to communities crime falls in those communities, and when the Government make welfare savings they are all going thieving; inquisitive crime goes up.

Professor Farrall: Yes, it is property crime. I am not sure I’d put it quite as crudely as that.

Steve Brine: Well, you did.

Professor Farrall: Well, no. I said that an increase in benefit spend is associated with a decrease in property crime. That is not necessarily the same-

Q276 Steve Brine: So areas of high benefit spend have areas of low crime.

Professor Farrall: Hold on. If you would let me finish, I am not saying there is a causal link. That is an association-not necessarily a causal link.

Steve Brine: Okay. It is a matter of debate.

Chair: I think Professor Laycock just wanted to mention something.

Q277 Steve Brine: Yes. Professor, I just want to ask you, with the work that you have done on community policing, presumably you have a view on this and crime falling.

Professor Laycock: I do have a view on the crime figures. I wanted to take a really radically different point in answering the question that you have raised and stick my neck out, if I may. I hope nobody cuts my head off.

Steve Brine: No, no. This is a neck-sticking-out kind of place.

Professor Laycock: The answer to "Why did it go up in the first place?", as far as I am concerned, is substantially because there was much more to steal. We had cars, televisions, mobile phones-all those things that we didn’t have in the 1950s when I was a child. There is far more mobility of young people, who are those at greatest risk of offending, and therefore far more supervision of them in those high risk years. Women go out to work now and I applaud that, being one of them, but it means that there is nobody at home guarding the house all day, so burglaries are a greater risk. That is, in my view, why it went up.

Why did it come down? It started to drop very rapidly in the 1990s. A very rapid change like that is not due to something like better parenting. It is just implausible. I can tell you chapter and verse on why there has been a 70% drop in car crime, and that is because in 1989 the UK had one of the highest rates of theft of and from vehicles in Europe. The Home Secretary called in the manufacturers and said, "Do something about the design of vehicles. You can open any Ford Cortina with any Ford Cortina key," and they said no, because they were not the victims of the crime and you bought another car when yours had been stolen. The Home Office published a car theft index which ranked the manufacturers by risk of having cars stolen. The manufacturers came into the Home Office, the Home secretary waved it and said, "Next time we publish this we will name the manufacturer at the top of this list", and they all went away and fitted deadlocks and immobilisers on vehicles. I can show you the figures. It takes 10 years to replace the car pool on the road, so after 10 years the dramatic drop starts to level off and that is what we have seen.

We have also seen a massive drop in what we used to call taking and driving away. If you remember, in the 1980s kids were driving like lunatics, stealing cars. That has virtually stopped. It is now very difficult to opportunistically steal a car, so there is a lot of evidence and it translates abroad as well.

With burglary, the research the ESRC has funded shows the same thing. The reason that burglary has gone down is that homes are more secure. The danger now is that the DCLG are proposing to drop the regulations on securing houses from their agenda because they want to make life easier for builders. If that happens, ACPO are predicting that the burglary figures will go up again. So the policy at national level is vitally important because it creates the context within which we can all live, and it can facilitate crime or it can make it more difficult.

Q278 Steve Brine: Okay, but what about violent crime?

Professor Laycock: Violent crime typically drops in a recession, or the rate of growth of it slows down. It is the other way round. Violent crime is also a very small proportion of all crime, don’t forget. The figures have changed. The Home Office counting rules changed quite dramatically so there was a spike in violent crime caused by a change in definition, which is a bit strange. Violent crime hits the headlines, which gives you a totally disproportionate view of its frequency, so we talk about knife crime and gun crime and so on. Those offences can be usefully dealt with. For example, in the United States there is some extremely good work on homicide-young people shooting each other. We can’t do it here because of the way our courts are organised.

Chair: Because of the way-?

Professor Laycock: The way our court system is organised. In the United States they have far more discretion at court level, so the district attorney can decide not to prosecute even though they’ve got the evidence. In the UK it is very difficult to do that. They use that evidence as a lever over the potential offender. "If you go out with a gun tomorrow, you will go to federal court and we will throw the book at you." We can’t do that here.

Q279 Steve Brine: Okay. Chair, Professor McDougall says she wants to come in. With your permission, can we just conclude this and then move on?

Chair: Yes.

Professor McDougall: I’d like to take a corrections view of this and talk about the positive things that we have done to combat crime because a lot of people have done a great deal to try and combat crime. You probably all know now, because it is very famous, that in 1974 Martinson wrote a report saying that nothing works; there is no point in doing anything; there is no point in probation; there is no point in interventions because people do not respond to this and nothing works with them. That was really grasped by all the countries, and in this country too we stopped doing interventions. We started thinking about humane containment in prisons. We just did that kind of thing. We did not do any interventions with prisoners at all, and this lasted for about nearly 20 years that we just did not try to do anything. Then, suddenly, researchers started saying, "This can’t be right," and they looked at all the data that Martinson had used, they did studies themselves, and they discovered and concluded that some things didn’t work-that is right-but some things did work if they were done with certain people in a certain kind of way.

This was international. It is not just in this country; it happened across the world. People suddenly started to say, "Okay, some things work and we will start implementing the things that work." It happened in the States and in Canada, it happened here, and it happened in Australia and in Europe-all the countries that started to listen and started to take notice of what was the right thing to do and what would reduce reoffending. That lasted for some time. We had an amazing system in this country where we had an accreditation committee that looked at all the interventions that were proposed; they trained people properly and funded people properly. This happened in other countries as well.

There was such a strong move towards reducing reoffending. Suddenly we all started thinking about it, and now, actually, it is very interesting that we are all in this Committee saying we are looking at one outcome-reducing reoffending. As to all the outputs that we previously used to ask for-counting how many times you saw people and so on-we don’t want that; we want to reduce reoffending.

So I would propose that this is bound to have an effect. It has happened in prisons and probation. They started to coordinate what they did. Instead of some people working with prisoners, saying one thing to them and other people when they were released saying something different, we all started saying the same thing. We all started training in the same way; we all started using the same philosophy and the same research evidence. I think that made an enormous difference to our crime across the world.

Q280 Mr Llwyd: Good morning. Professor Laycock, in 2001, in a speech which you gave, you mentioned-and I will quote: "The research literature on crime and criminals is packed with good ideas. Yet the discussion of crime remains doggedly based on intuition, anecdote, received wisdom and untutored opinion. The person on the Clapham omnibus is more likely to direct government policy than is the scientist." I don’t know whether the person on the Clapham omnibus is still alive but he must be getting on now, I should think. What I would really like to know from you and from all of you-I don’t know what the collective noun for four professors is-is, to what extent is Government policy on crime reduction being informed by research evidence, and can you give any example where this has occurred positively in the recent few years?

Professor Laycock: I still hold to it. I was a bit brave, I think, saying what I said, but I think it is still true. If you have a dinner party, the world and his wife at your dinner party will have a view on what you do about crime. If you say you have cancer, they won’t have a clue; they will defer to the medical profession on that. That is the sort of background to where I came from with the Clapham omnibus man. There is a lot in the research literature, and that is the purpose of the What Works research that we are now doing, funded by the ESRC and the College of Policing-to trawl that out and present it in a way that is far more understandable to practitioners. I do not, when I say that, mean that they are not capable of understanding complex things. Some academics are particularly bad at communicating their research results; they are bogged down in statistics and caveats and so on, and they are not clear about what the results might be. So it is a twoway street. There is a lot there that could be done.

My example of car crime is, I think, a very good example of where-I didn’t bore you with the theory behind it, but it is about leverage. If you have an agency that is competent to change the design of something and it doesn’t, it is probably because they are not paying the price of crime. For example, credit card fraud was very high until Mike Levy pointed out to the banks that if they didn’t do something, in the 1980s and ’90s, they would be losing a quarter of a billion pounds a year. Unlike the manufacturers of cars, the banks got together and worked out what they were going to do about credit card fraud. That was facilitated by research that the Home Office funded, so there are big high-level things that can be done by Government to create a context which makes crime less likely. There are examples where they have done that, like the banking fraud issue and car crime, and there are examples where they manifestly haven’t. I would take alcohol policy as my favourite example of that, where there is a mass of research literature totally ignored by Government.

Q281 Mr Llwyd: Some of us, over the years, have been arguing, for example, on what you have just said-alcopops advertising and so on. We were constantly told that the Russell Group have it in hand; they are dealing with it.

Chair: It is Portman.

Mr Llwyd: I beg your pardon-it is the Portman Group.

Professor Laycock: The Russell Group is universities.

Mr Llwyd: It is all these universities in front of me.

Professor Laycock: I am going to go back to UCL and have a word about this.

Mr Llwyd: If they were selfregulating, it was perfectly fine, but, clearly, that is not the case, is it?

Professor Laycock: This is one of the examples, actually, of what doesn’t work. There has been a really good book written in 2003 by Babor and colleagues called "Alcohol: No Ordinary Commodity." At the back of that book they summarise what we know about what works in terms of controlling alcohol use. Advertising is a very weak lever. The most powerful way to reduce the consumption of alcohol is to take it out of supermarkets, corner shops, sweet shops, every other which place, and put it back in off-licences that are properly regulated. The Government will not do that, for fairly obvious reasons. There is a massive lobby. Even increasing the tax-it did not happen. Increasing price is another way of reducing consumption, but it doesn’t happen because of the lobby.

On the question about the extent to which research influences policy, research is only one of umpteen things that influence policy. I don’t blame the Prime Minister for wanting to be reelected, for example. That is what politicians seem to want. It is perfectly rational. I don’t say it critically, but research, precedent, cost, politics, the media-all those things- influence policy. I would like to see research pushed up that agenda, but that means academics have to get better at presenting the research, and it might mean, sadly, that we have to lobby. I don’t like the idea but that is probably what we will have to do more of.

Q282 Mr Llwyd: Is that the view of the panel?

Professor Farrington: I wasn’t going to address that question but to talk about ways in which research has influenced policy. There are many examples, and I was involved in the Action Plan on Social Exclusion, which was the Prime Minister’s action plan launched in 2006, which was a five-year plan. This was an excellent plan which was based very much on research and it involved a lot of initiatives in the way of parent training, setting up a national academy of parenting practitioners. It involved using programmes which had been shown to be effective such as the Nurse Family Partnership programme and the Multisystemic Therapy and Multidimensional Treatment Foster Care. The Action Plan on Social Exclusion was a major initiative. It lasted five years-unfortunately no longer continuing because of the change of Government-but it is an example of a Government policy which was very solidly based in research and was very good indeed.

Let me give you one other example. My colleague Larry Sherman in Cambridge did a whole series of experiments on restorative justice, which showed that it was effective. Indeed, there was a cost-benefit analysis by Joanna Shapland from Sheffield, which concluded that £8 was saved for every £1 invested in restorative justice. This has very much influenced the Crime and Courts Act 2013. The Home Secretary has approved several million pounds of funding for police and crime commissioners to pay for restorative justice. Here, a series of experiments were carried out by Larry Sherman, which have had a major impact on policy in this country in the way of adopting restorative justice. I could quote other examples but I don’t want to go on.

Q283 Mr Llwyd: I am sure, Professor, that is absolutely right, but of course, restorative justice has been a core item in Northern Ireland for some time. So I dare say that some of the statistics from Northern Ireland have fed into the mix as well. Could I ask you then, on the other side of this particular coin, if there are any examples where policy seemed to be running, as it were, counter to the evidence on crime reduction or where policy could be more effective if evidence and data were better used?

Professor Farrington: I am sorry, I don’t want to be hogging the floor, but I would like to answer that in terms of what is going on in the Washington State Institute for Public Policy, because Steve Aos and his colleagues systematically reviewed what we know about the costs and benefits of different programmes, ranging all the way from social and developmental programmes, police programmes, situational programmes, correctional programmes, community programmes and so on. He made recommendations to the Washington State legislature about how best to invest money to save crime. As a result of all his reviews, the Washington State legislature decided to divert money away from imprisonment and towards early intervention. What has happened in Washington State since this initiative in the last few years is that you can see a decline in crime and a decline in imprisonment; so the two things went down together and they saved a lot of money. This is what we should be doing in this country. Happily, we are because the Youth Justice Board has contracted with the Dartington Social Research Unit-a man called Michael Little-to work with Steve Aos to try to produce a similar system in this country of trying to assess the costs and benefits of programmes and trying to make recommendations, initially to the Youth Justice Board, about where the best bang for the buck is: where you should invest money to save money.

The other thing I would say about this is that, when we talk about diverting money from imprisonment to early intervention, people say the public want money spent on imprisonment; the public are punitive. But if you actually do surveys to ask the public, that is not true. If you give them a choice, and if you say, "Would you prefer youth programmes, or would you prefer more imprisonment", or "Would you prefer nurse family programmes, or more imprisonment?", they normally choose more-they don’t choose imprisonment. If you actually do surveys where you ask the public whether they want more imprisonment or whether they want something else which is, if you like, more preventive or more enlightened, the public will actually prefer the more enlightened approach; they won’t prefer imprisonment. I think it would be very good for us to pay more attention to what has happened in Washington State, to carefully calculate the costs and benefits of all sorts of different ways of reducing crime, and to make rational recommendations about where to get the best bang for the buck.

Chair: It is perhaps worth saying that the Committee, in its previous composition, did look at Washington State when writing its previous report-the "Justice Reinvestment" report-and that drew significantly on the initial stages of what you have described.

Q284 Mr Llwyd: And, of course, when we visited Texas this year we saw a lot of good work being done there as well, avoiding incarceration and so on, with early intervention.

My final question, if I may, is to what extent is evidence on crime reduction well publicised or well understood by the Government, local policy makers, agencies and practitioners?

Professor McDougall: I think the academics have a responsibility here. There is evidence that Government-and this Committee is an example of that-want to know about research evidence. They want clear messages about what works and what doesn’t work. Unfortunately, the academic community doesn’t always agree. I agree with Gloria: we need to communicate that information more clearly, but we need to decide how we are going to explain it as well, because I can understand how people are confused about the research evidence because they hear different things from different academics at different times. We have different views on this panel-a well-chosen panel, I have to say, because we all have different views about things. We really need to make it clearer where we are coming from, because we have sociological approaches, economic approaches, psychological approaches and longitudinal approaches. They all go together, but the way they are communicated, it sounds as if they don’t.

We have lots of ideas and good theories, but that is not research evidence. That is something different. Yes, we want to hear what these ideas and these theories are, but we also want to test them and we want to do a fair test; so we want also experimental research to communicate that and find out what the results are. We want economic assessments, but we need to have good research on which to base the economic assessments, and we need good economic quality as well of assessment.

All of these things tie together, but I don’t think we explain it in that way and I don’t think we explain it to ourselves in that way actually. We are saying, yes, there are all these wonderful theories; we think these things will work, we are trying them and so on, but we should be evaluating them rigorously as we go along, in the same way as NICE does, for example, as the health services do, to test whether the good idea actually works or doesn’t work. We need to get our act together in that sense as well.

Q285 Mr Llwyd: Professor Farrall?

Professor Farrall: I think my colleagues are perfectly right. Academics like writing books and journal articles, which very few people read. One of the things which they need to do is to find other media in order to express their ideas. One of the things which I did with colleagues in Glasgow and Queen’s University Belfast, again funded by the ESRC-I have left a copy of it with the Chair-was to make a film about why people stop offending. We took a guy who now works as a criminal justice social worker in Scotland, but who himself had been involved in crime and been through prison, on a kind of "Who Do You Think You Are?" journey, in which he went to the States and spoke to people there, spoke to colleagues in the UK who work in prisons and probation, academics and policy makers, and people who themselves have been through the prison system, in order to put together an understanding of how it is and why it is that people stop offending. That was made with a professional film company, with the idea of showing it to practitioners, to give them a better understanding of the ideas around the concept of desistance. It has also been used in practice, not just in England and Wales but across the world.

Whenever I talk at probation trusts up and down the country I am always amazed at people who come up and say, "Yes, we have shown your film to people in Jamaica." They all struggled with the thick Scottish accent, but, other than that, it is those kinds of vehicles that academics need to get better at using. It won’t or might not surprise you to learn that, with regard to the grant which we have just started on, on Thatcherism, we have money to make a film for that because that is a really good vehicle. It is 30 to 45 minutes long and it is quite punchy. You do it with a professional company so you get your messages across clearly. Lots of stuff ends up on the cutting-room floor, but that is okay. To reinforce what Cynthia and Gloria have said, we academics need to get better at doing nontraditional civic engagement.

Q286 Mr Llwyd: Going back to pure research-Professor McDougall may have been a bit too hard on herself there-your research papers are all peer-reviewed, aren’t they?

Professor Farrall: By and large, yes.

Q287 Mr Llwyd: Really, it is up to people like us to read these peer-reviewed papers if we want to-or is it?

Professor Laycock: I wouldn’t recommend it, no.

Mr Llwyd: Or be given an understandable copy thereof.

Professor Laycock: It is for us to summarise those papers into packages for you. To answer your question, the reason I like science-well, the thing scientists do is, first of all, they are very clear on what the problem is, then they hypothesise about the solution, and then they test it. That is what we need to do more of. If the problem is how you make crime go down and you are worried about cost-effectiveness, in my opinion the best way to do that is to look to the situation, because the immediate situation determines human behaviour. In other words, opportunities cause crime. Imagine we took away all the locks and bolts and things. If you think crime would go up as a consequence, then you believe that opportunities cause crime, and I think most people would expect crime to go up if we took all our locks and bolts away. I am not saying we want to live in Fortress Britain; don’t get me wrong. It is about getting the balance right.

Q288 Mr Llwyd: As with the car crime?

Professor Laycock: As with car crime, but also as with alcohol policy and as with all these other major Government policies. When you have done the best you can on designing cars, you are then down to the local level to get car parking right and all the rest of it, and to get people to make sure they lock their cars and so on. That is fair enough to delegate, but you don’t delegate responsibility for alcohol control when you still have this ridiculous system. That doesn’t mean that I don’t think we should be offering treatment programmes for offenders. We are in a civilised society and we totally should, because that is what civilised people should do. I also think we should be educating our children as well as we can and supporting parents because we are civilised, not because if we don’t we might have a crime problem. In other words, it is the reason. Implying that we are educating our children because we are worried about our crime problem is a bit kind of hedonistic. It is not the kind of society I would like to be in. I want to educate children because they have a right to a good education.

Q289 Chair: Can I ask another question at this point? The Government are trying to incentivise and draw the voluntary and private sectors further into stopping reoffending. I don’t want to get into the argument of principle around that, but to establish whether the kind of process we are talking about here, doing innovative things but evaluating them and deciding whether they work and then moving on, is a process which can readily happen under this kind of system. Is it more likely to be feasible under this kind of system, or less likely?

Professor Farrall: I would imagine there might be situations in which it would be less likely to be feasible. If data on reconviction and reoffending rate are now, if you like, something which an organisation has a commercial interest in, then I am not entirely sure that they would necessarily want their rates of reconviction too widely publicised at some level. I am not entirely sure, if you like, of the level of ownership of the data in terms of reconviction, but I could imagine-

Q290 Chair: But that is something that could be got right; that is about how much information.

Professor Farrall: Yes, it could be something that is got right if you regulate it in a particular way. The data you regulate are only as good as the data that are reported to you. If people have a vested interest in making it look like it has gone up or look like it has gone down-and people may have vested interests in one or other, or in some instances both-then you can see obvious drivers towards what might be referred to as data manipulation. There is probably enough evidence in the recent past in criminal justice to suggest that that has happened and I can’t see any reason why it shouldn’t happen again.

Chair: In that case I will turn to Mr McDonnell.

Q291 John McDonnell: I was going to ask you a whole series of questions about what crime science is-but you have just demonstrated it, basically-about bringing it together and asking for examples of it. You have done that perfectly well. Can I ask you a general question and then get down to a couple of specifics? How do you think we can move forward then, just going back to the previous discussion you had? How can we move forward, both in terms of Government and politics generally, in getting a more coherent, evidence-based approach in this area, both in terms of what you can do as academics and also what Government can do?

Professor Laycock: One of the problems, I think, is that every time there is a change of Government there seems to be a change of policy. David sort of referred to it earlier: "Oh, that’s what the last lot did. Now we are going to do something else." For example, when the Labour Government came in in ’97 or whenever it was, they had the Crime Reduction Programme. That was a quarter of a billion pounds into crime reduction. I was in the Home Office at the time. I was speechless. We’d never had that much money. David Omand described it as half an aircraft wing because he had come from the Ministry of Defence. It was an extraordinary amount of money. The problem was we got too much too fast. We couldn’t spend it sensibly. Then the Treasury had this annuality nonsense that, if you did not spend it, you lost it, so there was a stampede to spend all this money. It was just nonsense. To have that kind of money but spread over maybe three or four Administrations would have been absolutely fantastic and we could have funded exactly this kind of research. I had £25 million for research and I hadn’t enough academics to give it to. It was just absurd.

So I don’t know how you get the consistency. Interestingly, we have it in relation to terrorism because all the parties agree that we can’t have terrorism, so you get a consistent policy across Administrations. We need the same attitude to crime. It is sufficiently serious. The public cares sufficiently about it, and cross-party consensus on what you are actually going to do is really important. I think they will all sign up to the idea of evidence because who is going to say they don’t want to know about what works? You couldn’t plausibly do that, I guess-well, maybe you could. I really am optimistic that the whole thrust now is of getting evidence-based everything. It has become the buzzword-evidence-based management, evidence-based medicine, evidence-based policing-but let’s hope we stick with it for a little bit longer than some of the Government ideas.

Professor McDougall: It needs to be independent, actually. That’s the point. The decision about the evidence base needs to be that the evidence has to be unbiased by political opinion, by information and current interest and so on. I know that is very difficult because it is a subject that, as Gloria says, everybody at a party has a view about-what you do with criminals and so on-and so you have to take public views into account. Nevertheless, now, in healthcare services, people now really, really accept that the evidence is important and, no matter what someone’s opinion is, you’ve got to go by the research evidence and follow that. We need to get into that situation with criminal justice but we are a long way off. Research in criminal justice is a disposable commodity.

I mentioned earlier the What Works ideas. We have had great success in developing What Works, and now we are about to throw it away because we are a bit bored with it and we will go on to another thing shortly. We don’t add desistance to what we are currently doing. We don’t behave like scientists and build up a body of knowledge; we chuck something out and start again with something else. It is the latest fashion and that really has to stop. It has to become very much a science and it has to build up the level of information, test it, discard the bits that don’t work through evidence, through evaluation, and add things that do, so that we end up with a very, very solid evidence base, which is not political, and politicians can defend themselves in making the decisions that they are being called on to do. The one thing that exists for ever is prisons. That is the one thing where everybody will say, "That’s great; we’ll always keep prisons," and yet we should be saying, "Let’s use prisons for the people who need to be there and use them less for the people who don’t need to be there."

Q292 Chair: Can I make it clear that the Committee, in its previous report, was seeking to do precisely that-that is, to move policy towards an evidence basis and to create some political space and consensus around the idea that that is what we should be doing? What we are looking at now is how far that was done. My own initial impression is that it did have some impact at the time, but it is always under pressure from some of the other political pressures that you describe.

Professor McDougall: There is always a military regime waiting round the corner to be imposed, isn’t there-short sharp shocks and detention centres?

Q293 John McDonnell: Stephen?

Professor Farrall: To take you back to your question about going forward, the answers are more primary research, better funding of primary research, and also funding projects that in academia we refer to as blue skies: fields which are, if you like, untouched or relatively unexplored or haven’t been explored for several decades-going back and seeing what can be learned or seeing what has changed, those sorts of things. Another thing that I would recommend-and I know that Gloria, in some respects, is a prime example of this, as indeed is Cynthia-is people who have, throughout their life courses, gone from being, if you like, Government or civil servant insiders to become academics and back and forth. Another example of that kind of career would be Nick Tilley’s, who is a very, very well respected academic but also has worked with civil servants, guiding not so much substantive knowledge, although that as well, but also thinking about research design in a very, very critical way that tries to get the best from all of the different research traditions, both quantitative and qualitative.

Professor Farrington: In terms of trying to influence Government policy it would be desirable for the Ministers to reach out to us, in a way. We can’t force ourselves on the Ministers. It is good when the Ministers ask us to give advice and we are only too ready to do that. Certainly, I have been involved with several Home Secretaries, including Kenneth Baker and Charles Clarke, and the Prime Minister’s Strategy Unit, as it used to be. It used to invite people to give talks and that was good. As I say, I was involved with the Action Plan on Social Exclusion. So sometimes the Ministers and the Prime Ministers ask for advice, and we are only too happy to give it to them and to try and summarise research for them. That is often more efficient for them than reading things because they are busy people and it is easier for them. If we can tell them something in half an hour, it is actually more effective than saying, "Read something." It is important to summarise knowledge for the Ministers but we need the civil servants to tell the Ministers to ask us, because otherwise we can’t get access to them very easily.

I would certainly say we need more primary research and, particularly, we need more randomised trials. It is absolutely ridiculous the few that we have had in England in the last 30 years. As an example, we have had no randomised trials in prisons for 30 years with recidivism outcomes. Cynthia did one but she hasn’t yet got recidivism outcomes, but there have been no randomised trials evaluating any sort of prison programme for 30 years with the recidivism outcomes. It’s madness. In medicine, we are doing randomised trials all the time to evaluate new drugs. In criminal justice, we are implementing new policies all the time and we are not evaluating them, so it’s no wonder we don’t know what works half the time. We need more primary research, really.

Q294 John McDonnell: Can I just follow that up? Are the structures in place in terms of Government engagement-feeding research into Government? Are there sufficient structures?

Professor McDougall: No.

Professor Laycock: No.

Professor Farrington: No, I’m not sure there are.

Q295 John McDonnell: It is the way in which information is communicated and properly commissioned-

Professor Farrington: It’s haphazard, I think.

Professor McDougall: This Committee does it. That’s about it.

Professor Farrington: It depends, often, on the civil servants advising the Ministers-so, call this person in or ask that person for evidence. It is very haphazard. There is no real system.

Professor Farrall: There is anecdotal-well, it’s not anecdotal, but it is patchy when it works. What universities have had to do over the last five or six years is get good at what we now term "impact," that is, changing the real world-for the better, I hasten to add. Therefore, universities are much more outward-looking in terms of disseminating their research findings. In some respects what we need is something at the other end that we can sort of plug into. We need an organisation, a branch of the civil service, even at Department level or across the whole thing, that actually looks away from Westminster and Whitehall out to universities, not just in England and Wales or the UK but across the globe-because we might have 20 or so top-notch universities in the Russell Group, but there are a whole load of other universities elsewhere that we should be talking to as well-that actually starts to try and bring academics in, because I think the academic community by and large want to engage. It is one of the reasons that people go into academia-because there is something they feel passionate about, that they want to know more about, that they want to have some kind of transformative effect on, and so to have somebody trying to pull you into Government is great.

Q296 John McDonnell: Professor Laycock, you have been in the Home Office. Has there ever been a model that you would recommend of engagement in that way?

Professor Laycock: A formal structure, I don’t think so, no. One of the problems with crime prevention is that it is cross-departmental. Some of my favourite examples of where we actually got something done is that it actually took the Prime Minister to get them done because Departments were arguing with each other. That is absolutely absurd.

Q297 Chair: We found, as a Committee, when we were in Germany a few years ago, that UK academic research was being quite widely used by Administrations there, and one or two other countries, and referred to. Of course, that did not depend on there being some formal structure of engagement between UK academics and foreign Governments; they simply saw that it was good research and made use of it.

Professor Laycock: We were talking outside about what had happened to the Home Office Research Unit, where the money had gone and why it wasn’t there any more. I was in it and I was also in the Police Research Group, which was independent of the Home Office Research Unit. With regard to the Police Research Group, I had more or less direct access to Ministers through the head of the police department and I could feed all the research straight through. All the work on repeat victimisation, for example, which was done, funded by the Home Office, went to Ministers. We were determined it was not going to go away and we made the police pay attention to it because we could. We had leverage over the police. That is not there any more. I don’t know how my colleagues feel about that, but I think it is a great loss.

Q298 Steve Brine: Professor Laycock, I am still reeling from your comment earlier that you had £25 million for research and "we couldn’t find enough academics to give it away to." No wonder this country is broke, but that is a political point.

Chair: For another day.

Steve Brine: For another day, yes; I’ve made it. Isn’t it better now that we have quality over quantity of research, in a time of restrained budgets? Doesn’t the taxpayer get better value for money?

Professor Laycock: I think we have gone far too far in the wrong direction. There is a balance here between absolutely nothing and what you might call too much too fast. There is nothing wrong with a quarter of a billion pound investment in how you control crime, but you can’t spend it within the time frame they expect you to spend it in. If you said £5 million a year on investing in crime, that is not a lot of money in Government’s terms at all.

Q299 Steve Brine: But when you had all that money and you said you couldn’t give it away quick enough-it’s like that film "Brewster’s Millions", isn’t it?-did you and Ministers not push back to No. 10 and No. 11 and say, "This is crazy"?

Professor Laycock: We did get permission from the Treasury to waive annuality, but, to be absolutely frank-if I’m going to be really frank with you-

Steve Brine: Be frank.

Professor Laycock: I was in research, pushing the research bit. There were policy makers who had control of the whole programme. I had to persuade them to persuade Ministers. I couldn’t go over their heads. You get shot for that and I wasn’t prepared to-it’s rude, apart from anything else.

Steve Brine: Of course.

Professor Laycock: There are all sorts of reasons. After six months I’d had enough and left and went to the United States. We have since written a book about it.

Professor Farrall: Gloria’s point was not that she had 25 million quid but she had 25 million quid and she had to spend it very, very quickly. It’s very hard to spend money very quickly.

Chair: I think we’ve got the point.

Q300 John McDonnell: Just to finalise that, I would welcome your ideas on how you would structure academic engagement with the Government in this area so you can have a planned approach and a consistent approach over time so you don’t go through this ricocheting.

Professor Farrall: I think of secondments. Universities are quite used to selling staff across universities, almost.

Professor Laycock: Nick Tilley was exactly that. He was in the Police Research Group from Nottingham Trent University and he was there for more or less 20 years.

John McDonnell: Any ideas that you have would be really, really helpful.

Professor McDougall: Can I say I think we have got a good research team in NOMS at the moment. They are producing wonderful stuff with, for example, the Offender Group Reconviction Scale and being able to say this is how we are going to assess risk. They are following instructions and answering questions about research. What they don’t have is a wide enough scope. They are working under a very narrow scope about particular questions that are being asked in NOMS. The difference with the Research Unit was that it could be wider, and it could go and seek out what is happening in universities and find out about that.

Q301 John McDonnell: It is the breadth across Departments, but I am also interested in what you think-don’t answer it now because we are obviously stuck for time-about how you get that political engagement. The point you make about the Prime Minister leading on some issues is fundamentally important, so from your experience-

Chair: Is this a question, Mr McDonnell?

John McDonnell: Yes, it is. I will come back to a specific question, which is drilling down into it. I am interested in the issue of desistance because you have dealt with some issues around the opportunity to commit crime. We also have to look at the motivation in some ways as well, so I just want to know what you think are the key factors in supporting offenders to stop committing crime.

Professor Farrall: Most of the evidence at the moment, I am afraid to say, provides answers which suggest you need to look away from the criminal justice system. One of the big relationships between involvement in crime is age. It is what we referred to as the peak age of reoffending-David contributed to this significantly, producing these kinds of, if you like, baseline findings. Generally speaking, between the ages of 10 and somewhere between 19 and 21, people’s involvement in crime is going up. Then, depending upon the nature of the offences they are committing, the period and the cohort that they are in will peak and then start to decline. It is in the kind of post-21 to 25 that we really start to see processes of desistance kick in. The factors which are most strongly associated with that are things like engagement in the labour market and in common social institutions like families. There is also, if you like, an interaction with age, so, if you are engaged in crime and get a job before around about the age of 21, that tends to actually accelerate or increase involvement in crime because money is used frivolously to buy drugs and drink and things like that. After 21 to 25 you see that process flip and it reverses, and you get a much stronger relationship between employment and desistance from crime. They are the three big ones: age, employment and marriage.

We have other things that are much softer processes, much more internalised. For example, finding the motivation, finding a reason to want to stop and believing that you can indeed stop are important factors or processes in driving people towards desistance from crime. For some groups who have offended, the desire to give something back is very prominent. I think this is true of almost all people but probably not true of absolutely all people, although it is painful to say it. People like to say sorry when they realise that they have done something wrong. It may take them a very long time to realise that they have done something wrong, but there is something good and restorative about saying sorry and then being reaccepted. People want to apologise when they have done wrong and we have to find avenues to enable people to do whatever they consider to be an apology, which may not be apologising to their individual victim, because that may be impossible or in some instances inappropriate, but some way of giving something back.

For some groups, one of the things that drives them out is fear of who they will become: that if they carry on down that route, not necessarily that they will be incarcerated-although there is evidence to suggest that some people think like that-but they will become somebody who is targeted by other people in the local community because they are a known fence or a known face or encroaching on someone’s turf, as it were.

The break-up of the peer group is another key factor, particularly for young men. There is moving away from home; again, this is another study that came out of David’s research. Those men who left London and went to live elsewhere were more likely to stop offending. The explanation, which I think is still the current thinking-David will, I am sure, correct me if I am wrong-is that, by moving home, these men severed that peer group and, also, they had a loss of the label of somebody who was involved in crime. They could start afresh, as it were.

In terms of what the criminal justice system can do to help people-I am sure Cynthia is much better able than I am to talk on this-the research which I have conducted on this cohort of 200 men and women was very depressing to start with because they said some frankly unrepeatable things about the probation supervision they were getting. They liked their probation officers but they thought it was just a complete waste of time.

That view largely persisted through the first three waves of interviewing, so that is during the first two years that they were on probation supervision. They always remained in the community, by and large, but, when we followed them up afterwards-probation had ceased and they were no longer being supervised-after about five or six years, you started to see, among some of the individuals, the recognition that they had taken something from probation supervision. Then we finished interviewing them for a fifth time about a year ago-it took about three years to interview all of those that we could find-and there we find far more people saying, "I have changed," which they could demonstrate in lots of different ways, but also that they were now acknowledging that the things that their probation officer had said to them 10, 12 or 13 years earlier actually had been the seed, the starting point, of why they had started to change, as it were. Our explanations of why people change need to become much, much more complex, much more nuanced and much more stretched over time.

What happens-and I have talked to probation staff about this-is that things that they say in supervision sessions get lodged in people’s brains; they just remember them. I do not think offenders are particularly different from other people. I can remember things that my parents, teachers and friends said to me which I have stored as good bits of advice. Lots of what probation is about is aspirational and, for the individuals who have been supervised, quite often the problems that they have to deal with are pressing and so their aspirations are, if you like, delayed. When you fast-forward over a few years, you then find individuals who are in circumstances or situations in which they are better able to start thinking about change, maybe because they have embarked on a relationship, or they have already had a child with somebody and they have started another relationship and that is going to produce a child. At that point they start to cast around for new ways of being that don’t involve crime, and they then start to draw down some of what they can recall that their probation officer had said to them in the past. The complexity of this is quite great. It is unpicking it just at an individual level.

The problem, therefore, for the criminal justice system is this. Probation staff get this; whether managers and people in policy share this view I am less certain of, but probation is an investment in the future and it is a bit like a mortgage or a pension scheme. You have to wait for it to mature. You are not going to get, in very many instances, dramatic turnarounds straight away. There will be some, but for those people-

Chair: We have to let somebody else contribute. This is a very full and-

Professor Farrall: I was going to say that those people who are very heavily entrenched will need to go through a very long process of change, and that takes time.

Chair: I think Professor Laycock wanted to say something.

Professor Laycock: Just very quickly, absolutely all those things are right: marriage, family being held together and so on. They are not helped by people being put in prison, partly because the prisons are a gazillion miles away from where these people live, so community prisons are a really sensible idea. Mega, great big prisons are a disastrous idea. Titan prisons have gone but the next biggest prison is also a bad idea. You want small, local prisons that facilitate ties with families, jobs and so on.

Professor Farrall: There are very few of those for women. There are none in Wales.

Q302 Chair: We debated our report on women in the criminal justice system only last Thursday. I really want to give Yasmin Qureshi a chance to come in.

Professor McDougall: Could I add something about the probation debate? It is important to say that probation has evolved as a profession and it is a mix of all kinds of different theories about how you do probation. Faye Taxman, in the States, says that it is completely atheoretical, and it is. It is built up with a lot of ideas of people with different kinds of learning that they have had over the years and so on. It has built up in that way. However, we know that we can make successful interventions if we do them in a structured way. Now, in the States, in Canada and certainly in the UK in small pockets, we have been looking at doing structured probation and being very clear what it is that we are doing. I have just been involved in an evaluation-a randomised control trial-which showed that this was effective with high-risk offenders. Structured probation was more effective than people who did not have that particular intervention. In Canada they have shown that, if you train probation officers to do the structured probation, and they randomly allocated offenders to those who were trained and those who were not trained, there was a big improvement in the offenders who were being supervised by the probation officers who had been trained. There is a lot of evidence to say you can make probation better or quicker, actually, rather than, as Stephen was saying, having to wait rather a long time for it to work. You can actually facilitate that process. That is what desistance is trying to do-to speed up the processes that would happen naturally, but we are trying to give it a bit of a hand.

Q303 Yasmin Qureshi: I just wanted to continue the discussions about cost-benefit analysis and the economics of it. I agree with you, Professor Laycock, when you say that we should not be thinking about the money aspect of it. Of course, at the same time what tends to happen is we know that certain sections of our media seem to suggest that the only way to deal with crime is to bang everybody up in prison for as long as possible and throw the keys away. I just want to say that I do not agree with that. I spent 20-odd years prosecuting and defending people charged with criminal offences, and I saw, in those years, casework from young children, the middle-aged, young adults-all sorts from different economic social backgrounds, charged with different types of offences. I just want to say that, although that might be anecdotal evidence-

Chair: I am losing track of the question, I am afraid.

Yasmin Qureshi: I agree with what you were saying that your research has shown, because it confirms a lot of what people like me would consider to be anecdotal evidence, just from my experiences. I wanted to set out my stall, as they say. Some people believe in this concept of throwing people into prisons, but what you all said is that a lot of policies have worked in different ways, whether it is the fact that cars are more difficult to steal now or hot wire, or burglary is more difficult because windows are now double-glazed, etc. You can’t just break a window and steal things, and there are also some of these intervention policies, but then people say this costs too much. That is one of the reasons some people say-

Chair: I am struggling, and I am sure our witnesses are, to find out what the question is.

Yasmin Qureshi: Just bear with me for one moment because I am just trying to explain my thinking and then the question will make sense, because if I just ask-

Chair: Could you please arrive at it fairly soon?

Yasmin Qureshi: Yes, I will; I will. Just bear me with me for a few seconds. I have just lost my train of thought there. Anyway, the intervention policies have been working.

Professor Farrington, you have given the example of Washington State, and you have, in very recent articles, talked about the issue of the cost-benefit analysis. Is there enough evidence in this country for the Government to be able to say, "Yes, we are happy to spend this money on interventions and all these policies, and at the end of the day we will get not only reduced crime rates but the cost of dealing with criminal activities-prosecuting, sending people to prison-will actually go down"?

Professor Farrington: You can’t draw that conclusion from the UK evidence. You would have to draw it from evidence in the United States, for example. We need our own evidence. We don’t have enough well-designed primary studies because any cost-benefit analysis is only as good as the primary study on which it is based. So it needs to be based on a high-quality evaluation. The answer is that in this country we don’t have enough primary knowledge, but, based on knowledge from other countries, we could certainly draw the conclusion that it would be better to invest in early intervention rather than imprisonment, for example. But I have to say we don’t have the knowledge in this country; we haven’t done enough of the primary research.

Q304 Yasmin Qureshi: Can I then ask all the professors this? Say you were given the money now to be able to carry out some of the research that you think needs to be carried out. What kind of research would you carry out? What would be the methodology in order to get the answers that you need, so that you can submit them to a Minister or Home Secretary and say, "This needs to be"-

Professor Farrington: I would embark on a programme of randomised trials to evaluate the impact of all sorts of interventions ranging from, for example, cognitive behavioural skills training programmes in prisons or probation, which we are using at the moment. Cynthia did a trial on it but with no recidivism data as yet. Basically, we have had no real evaluation of that. We could have more randomised trials on parent training. We have had some of those but not too many. We could have more randomised trials in court on, let us say, fines versus imprisonment. We have had some randomised trials in courts, such as restorative justice, but we could have more fundamental randomised trials in terms of what the relative effectiveness is of community penalties and fines. We don’t know that. There are all sorts of things.

The randomised trials we have had have tended to look at new innovations like restorative justice. They have not looked at the things we have been doing for years such as prison and probation and fines, and all these other things. If I had lots of money, I would embark on a whole connected series of randomised trials to evaluate the things we are doing now and also things we might do in the future.

Q305 Chair: Forgive me for intruding, but for the things you have described we have existing information. We know what the consequences have been of fining people and imprisoning people.

Professor Farrington: We do not know what the effect of fines versus imprisonment on reoffending is.

Q306 Chair: We can look at the people who were fined and the people who were imprisoned.

Professor Farrington: But there could be preexisting differences between them that explain any subsequent differences. The beauty of the randomised trial is that, because you randomise, the people who get one treatment are exactly the same as the people who get another treatment, on average. If we just follow up people now who get fines or imprisonment or community penalties, they could have been different to start with and that could explain any subsequent differences. So we need the randomised trials, as in medicine, to take out the preexisting differences.

Q307 Chair: What you can’t do is organise randomised trials by giving people sentences which the court doesn’t think they ought to have.

Professor Farrington: In the past this has been done where you give people something better than they would get otherwise. For example, when community service came in for the first time it was presented as an alternative to prison. You could easily have then said, "Okay, let’s take some people who would normally get prison and let’s randomly allocate some of them to get community service." Then it would be a test of whether community service or imprisonment was the better treatment. In a way, nobody was being penalised because they would have got prison and so some of them were getting something better. That is a way you can do these trials and it has been done in other countries.

Professor Laycock: I would spend some money evaluating problem-solving courts. As I said earlier, I am very keen on problem solving. Right now people come up before the court, the judge has this string of disposals and he announces what he announces.

Q308 Chair: Forgive me-again, hasn’t that been done in north Liverpool?

Professor Laycock: It needs to be tested more widely. It has been done in relation to drugs courts, for example; domestic violence courts have been looked at, but it is the whole notion of "What’s the problem here?" I mean, assessing the problem in a very wide sense. Is the problem that it is just so easy for this person to carry on committing crime? Is the problem that he has no alternative? Is the problem that he is taking drugs? What are we going to do about it? There has to be a consequence for that individual, so he has to come back to the court and there has to be a treatment programme out there that he can be frogmarched to and start the next day. We do not have that in place. There is a queue to get on a drugs treatment programme.

Q309 Yasmin Qureshi: Just to continue, you have mentioned about different measures of early interventions or intervention methods to help people committing crimes. Has there been a detailed study done on the fact that early-years intervention-i.e. the three or four-year-olds who come from broken homes or are socioeconomically disadvantaged-could actually prevent a whole series of criminal activities later on?

Professor Farrington: We have had quite a few randomised trials on that in different countries, including some in this country, which show that the Nurse Family Partnership, which involves giving advice to mothers in the first years of life or in pregnancy about how to bring up children, about infant development, avoiding smoking, drinking and drug use and such like, is effective. We know that high-quality pre-school education and pre-school intellectual enrichment programmes are effective. We know that parent training is effective. There are a lot of things that we can do in the first five years of life which we know reduce offending later on in life. We have had quite a few long-term followups which show that, plus we have had quite a few cost-benefit analyses based on them. In the Perry Program, for example, which is a pre-school intellectual enrichment programme, the latest claim was that $17 were saved for every dollar that was expended on this programme when you followed up to age 40. It was a programme given at ages three and four. There is lots of evidence, again mainly from other countries but some from this country. We have had some trials of early parent training in this country, and we have had some trials as part of the Action Plan on Social Exclusion, on Nurse Family Partnerships and other related things.

Q310 Yasmin Qureshi: Some of these projects have now dwindled, have they not, or are they still-

Professor Farrington: As I say, the Action Plan on Social Exclusion was only for five years and it stopped in March 2012. I don’t know whether it will continue. I used to be on the advisory board of the National Academy for Parenting Practitioners, but that seemed to close down in 2012. So all these things have a limited life. In our book on "Saving Children from a Life of Crime" we recommended the need for a national crime prevention council in this country, modelled on Sweden and Canada, where they do have national crime prevention councils. This could give advice to the Government about effective programmes, but, more importantly or equally importantly, it could give advice to local people because the problem is, when local people sit around and discuss, typically they lack knowledge and they need technical assistance in knowing what are effective programmes. Also, more importantly, a national council could provide continuity over time. You would not have this business of programmes starting and then finishing. If you had a national council, it could do a lot of things which would greatly help this country in having a sensible programme of crime prevention, crime reduction over time, helping the Government upwards and helping the local communities downwards.

Professor Laycock: One of the six What Works centres the Government is funding is the Early Intervention Foundation, which is looking at all sorts of effects of early intervention. There will be a crime element to that, I suspect. I totally agree with David’s point about the need for a national crime prevention council. We have one, then we don’t have one; and we have one and then we don’t have one. We used to have an inter-ministerial group on crime prevention and that fell by the wayside. It needs to be independent of Government but an advisory group, because then it can be cross-departmental and independent, and it will outlive the political-

Chair: In which case I will turn to Mr McDonald.

Q311 Andy McDonald: Thank you, Sir Alan.

You rather called the point that I wanted to address. It is interesting that everybody can have a comment and an opinion about crime reduction, as you quite rightly said. Sometimes the media has a lot to do here because it is an easy issue to chip in on and express a view. The Committee, in 2010, recommended an independent cross-disciplinary national crime reduction centre, much akin to NICE. We now have the What Works Centre, we have the Justice Data Lab and, of course, the Probation Institute. Interestingly, that is established next month or in March, and it is to safeguard professional standards and act as a centre of excellence. In terms of the standards of excellence, rigorous standards, the assessment of research and other evidence and its implications for the delivery of service protects the public and rehabilitates offenders. We have several bodies all pointing at the same thing. I know we have addressed it to some degree, but are we saying that the best way forward is to bring everything together in one singular centre of excellence that is going to inform us? Is that our collective view?

Professor Farrall: My thoughts are that it is great having lots of people interested in the same thing in the same place down the corridor, which is why I like working at the University of Sheffield because we have all of that there. The problem is that, if it is all corralled into just one place, you run the risk-I am not saying that this would happen-of having a homogeneity of thought in which only one model of research is considered to be sufficiently robust, or only one theoretical approach is ever properly dealt with. Just to chip in at the beginning, that is all laudable, but there are some things which you have to ensure don’t become embedded in it. Plurality of approaches and plurality of thought and an open-mindedness are one of the things which academics guard fiercely-and rightly so. I think that that should work almost more so in policy-related environments.

Professor McDougall: David and I would probably disagree with that because we think that there are very robust methods-experimental methods-that are recognised as being scientific and should be used. There is a large body of people who don’t want to use them and I am not quite sure why. There are a lot of reasons given, and a lot of practical reasons why randomised control trials won’t be used, but I sometimes think they are not used because they don’t give the answers people want. Randomised control trials are sometimes very, very unforgiving. People have a pet theory of what they want to do when they do a randomised control trial and it says it doesn’t make any difference. That is the problem. Some of us are committed to saying we need to know that; we need to know if things don’t work; we need to face up to it and we don’t need to be emotional about the methods that we choose to use. They have got to be methods that work, but a lot of people don’t agree with that. Therefore, we will have a problem, because with research evidence that is presented it can be of different standards. If we have a bar lower than the international community of researchers accept, then we would be ignored, really. The UK has a lower standard, in some senses, than other countries.

Professor Laycock: My comments are against the backdrop of directing the What Works crime reduction research to support the What Works Centre in crime reduction, which is based at the College of Policing. I share Stephen’s concerns about the danger of having the ideas in one place, and particularly something about crime reduction in the College of Policing because it is necessarily going to skew the work towards policing; so it is important that we make sure that doesn’t happen.

On the point that Cynthia makes and feels David shares about randomised controls trials, my feeling is if you want to know what works you need to know three things. You need to know that the methodology that determined the outcome of the experiment is appropriate and proper, and it has to be appropriate to test the hypothesis that you want to test. That may or may not be a randomised control trial. It depends. You need to know the mechanism; in other words, you need to know how did it work? You need to know the context; in other words, in what context does it work? For example, CCTV might work great in car parks to stop car theft, but it will do absolutely nothing to stop antisocial behaviour on a Friday night because everybody is drunk. It just won’t work there. You need to know the context, you need to know the mechanism and you need some reassurance about the methodology. Those three things are necessary. Just the methodology alone is not sufficient.

I will give you a very quick example. There was research on mandatory arrests for domestic violence in the United States, a project that Larry Sherman did: randomised control trial, mandatory arrest, best way to deal with domestic violence, roll it out across America. Three attempts at replication failed to replicate it. Why? Because, in some areas, if you arrest the offender and he already has a criminal record, the arrest means nothing to him and he goes home and beats his wife up even more for calling the police. In other words, mandatory arrest works best for people in a job-"middle-classy" people who care about being arrested because they are embarrassed by it, and you will see what the mechanism is. In that context it works, but it doesn’t work everywhere for everyone and the research needs to tease that out.

My phrase at the moment is that senior police officers, for example, need to have evidence-based judgment. They are still going to have to make judgments. It is not about looking at tick boxes on a "what works best" kind of Which? report. It is choosing your judgment but informed by the best possible evidence.

Professor McDougall: Well, professional judgment is the worst method of determining it.

Professor Laycock: I didn’t say only professional judgment-I said informed by the best possible judgment.

Professor McDougall: Even then, it is-

Professor Farrall: My point is that I suppose you don’t want only randomised control trials. They are fine, but some of David’s best work has been his longitudinal study, which is not a randomised control trial. There is qualitative research that has brought all sorts of things out. Ethnographies, historical research-all of these things have a part to play. My warning was only a warning. It is a caveat that all of these methodologies should be embraced and used because all of them have advantages, but they all have flaws.

Chair: We are starting to run a bit short of time, so I think we should move on.

Professor Farrington: I would like to answer your question, which is, "Should we have a national centre for crime prevention?" I think we should. As I say, I have said it before because it can be a centre of expertise, it can advise the Government, it can provide technical assistance to the local people who need it. We have two models in the world in Sweden and Canada, where they have a national crime prevention centre and this feeds lots of information to local crime prevention centres. We could build on our Crime and Disorder Reduction Partnerships, which we have in all the local authorities, and we can have a national centre which would advise them on how to more effectively reduce crime in their communities, because at the moment they don’t have the knowledge that they need.

Q312 Chair: We have that point as well. I just want to clarify the value or otherwise of the Justice Data Lab. Does anyone have any experience of that?

Professor Farrington: My understanding is that it is going to encourage people to use PNC data. The PNC data which is currently given out by the Home Office and the Ministry of Justice is unsatisfactory because it is very difficult to work out if the person who they send you is the person that you are trying to find. It is okay if you are following up ex-prisoners who have a PNC number because then they can be linked with a number, but, if you are searching a sample of, say, 500 people from the community and you want to know how many of them have a criminal record, very often there is no way of knowing whether the person they give you is the real person that you are looking for. So we need improvements in the PNC data that we have now if this is going to be the basis of a data lab because we need to be sure that they will give you back the person that you are looking for.

Q313 Chair: Isn’t there more generally a problem that sentencers don’t have this knowledge of the outcomes of their sentences?

Professor Farrington: That is true, yes.

Q314 Chair: Have any of you been involved in talking to sentencers, conferences for judges and magistrates and so forth?

Professor McDougall: Yes. They certainly don’t know the outcome. There is no feedback on that. They do not know what works with the offenders or what doesn’t work with the offenders. Similarly, they do not think about the costs and benefits either. They don’t think of costs of a sentence; they don’t think of what it costs to send somebody to prison. In fact, maybe committing a crime would be much less of a problem than the cost of running a prison sentence. So they never look at the costs and benefits of the sentence that they are giving.

Professor Farrall: I must say I had not heard of the Justice Data Lab until I spoke to one of your colleagues on Friday, which I am sure is my failing and not theirs. When I looked at it I got the sense that it was entirely quantitative data. That is fine, but there are a whole load of other data sources that should be used as well. I also got the impression that the data that it had related largely to what the criminal justice system knew about itself and what it did to people. Therefore, they don’t include any of those things that we refer to as dynamic factors, like whether somebody has a job or got married-all those sorts of things.

To go back to Yasmin’s question about what would I spend my 25 million quid on, this is what I would do. When an individual is sitting with a member of probation staff who is assessing them for a sentence plan, they fill in a form. That form is negatively loaded so that if something is not a problem it is zero and if it is a problem it is 10. That way of thinking constructs that individual sitting in front of you as, if you like, the physical embodiment of a whole series of social and economic problems. What it also needs to do is to have a way of capturing the strengths that that individual has. The criminal justice system sees people as problems. They are problems; that is why they are there in front of you. But individuals also have strengths, or at least they may have strengths or things that you can develop. We need a criminal justice system that accepts that individuals have deficiencies but also accepts that there may be strengths that can be built upon. It could be a relationship or a particular interest, and it is that thing that probation staff or people working in prisons need to latch on to. You need to find a reason for somebody to stay out of the system. Continually just focusing on their deficiencies just reminds individuals of their own failings and that is not going to bring them away from crime.

Q315 Steve Brine: Some of this has been touched on, so we will be brief here. I just want to turn to local approaches to crime reduction. As part of this inquiry, as you would expect, we had some of the police and crime commissioners in. A former member of this House, Alun Michael, who is now a PCC in South Wales, was one of those who said-and I can almost feel Professor Laycock sighing as I say this-that they wished to take an evidence-based approach. You would be surprised if they did not say that. He was specifically referring to reducing alcohol-related violence using A and E data.

Looking at the way that policy has moved since 2010 with respect to the creation of the police and crime commissioners-there is a question mark over how long they will be there-Health and Wellbeing Boards, and now the new providers of probation services, the CRCs, do you believe that there is sufficient evidence to inform the PCCs, the Health and Wellbeing Boards and the new CRCs to commission local services on a very local level? Who wants to dare to start on that one?

Professor Laycock: I really don’t think there is. It is terribly difficult to generalise. Some of them are ex-chief superintendents; some of them have a business background; some of them don’t know anything about crime; some of them used to be involved in the police authority. So it is terribly difficult to generalise. Commissioning, in effect, research, which is what we are all advocating, notwithstanding our minor disagreements about how you do it, is actually a skill in itself. The PCCs need some advice on how to commission that and how to ask the right questions of some of the demands being made by the police.

For example, the PCC in Sussex, whom I was speaking to recently, said she had been asked by the police to fund ANPR cameras in Brighton to stop burglary. She said, "How will that happen? How will it work?" It was exactly the right question.

Steve Brine: I think she was here, wasn’t she? We discussed this.

Professor Laycock: Yes-Katy Bourne. It was a really, really good question. They need to ask that kind of question-how will that work?-before they go doing it. Then, as I said to her when we were talking about it, do it experimentally and see if it works, but I can tell you right now just-this is why it is important to know about "how" and "context". I wouldn’t put money on that particular experiment, personally.

Q316 Steve Brine: Have any of the other members of the panel had any experiences of PCCs, with similar examples to Professor Laycock’s, which would be useful for us in putting our report together?

Professor McDougall: Not of PCCs but of just ad hoc approaches from charitable agencies, community support agencies, which come and say, "Please can you tell me what I should do? Somebody knows your name and maybe you can help and advise us, because we think we should be looking at whether we are effective or not because we are going to be paid by results. What do we do?" That is by chance that they come. They look you up on the internet or somebody tells you their wife knew you before, and things like that. So it just seems to happen in a very, very ad hoc fashion and it does not seem to be sufficient.

It seems to me, again, that we can learn such a lot from the healthcare services, because they have a centre of reviews and dissemination-at York University actually. All the information is available there on all the different interventions, and people can look at that for free and say, "What do you do for ingrowing toenails?", or something like that, and there is all the information, the best information, and it is provided. We need some kind of centre of reviews and dissemination for criminal justice interventions and things to do, things that would be helpful and that will help with research.

Q317 Steve Brine: For PCCs that has been difficult because they have not been a body. They have been a set of disparate people in their areas, which in some ways is the policy intention, but they have now started to come together as a group with more of a national association.

Professor Farrall: I have not had very much contact with PCCs, but I was going to say that I suspect they are moving in the direction to be more interested in research. Sheffield is one of eight institutions known as the N8, which is a subdivision of the Russell Group, which has recently started work with PCCs, looking at a whole range of policing issues. If you would like more information on it, the person to contact, who is leading the programme, is at Leeds University and his name is Adam Crawford.

Chair: We have a couple of quick supplementary points from Mr Corbyn first.

Q318 Jeremy Corbyn: Thanks for the evidence this morning. I represent an inner-city area, and every couple of months I have a meeting with the police and we go through the crime hotspots map, etc., etc. I know this map very well because it does not change very often. In your research, how much do you factor in, obviously, poverty levels in particular areas of high crime? Also, what bedevils my community perhaps more than most is an enormous population turnover. We are looking at 35% a year in some wards. Do you factor this in in your crime reduction ideas?

Professor Laycock: I think it is part of the reason that that is a hotspot, because you need three things for a crime: you need a motivated offender, a vulnerable victim or object or something, and the absence of capable guardians. In my area we kind of know each other, but where there is a very high turnover people do not know each other and there is a certain anonymity of people there-young people in particular-so they feel that they can offend with impunity because they aren’t going to be caught, as it were.

Q319 Jeremy Corbyn: Have you robustly looked at the neighbourhood watch schemes? Do they actually make much difference? Are they effective?

Professor Laycock: I was once asked that question by the Home Secretary’s secretary. She wanted an answer now and she wanted a yes or a no. I ended up saying yes, but I said yes because I believe the mechanism works. If offenders believe that, if they are seen committing a burglary someone will phone the police, that will put them off. The problem in your area, first of all, is that the offenders don’t believe anyone is going to see them and they don’t believe, if they do see them, that anyone is going to phone the police. In other words, in inner-city high-crime areas neighbourhood watch can’t be implemented. In my area it can be because we all phone the police if we see a burglary in progress, whether we have neighbourhood watch or not. But if that is the mechanism, if that is how it is supposed to work, then it depends on the context. The big challenge in a high-crime area like yours is to implement it.

It was done in relation to repeat victimisation on the Kirkholt Estate, very famously. What Ken Pease did was to take every known victim of burglary and ask their immediate next-door neighbours to wrap a mini-cocoon around them and they called it Cocoon Watch. "Watch out for them; they might be burgled again," and they introduced the neighbours to each other. Every time there was a burglary on this incredibly high-crime estate they had another Cocoon Watch. After 12 months they launched a proper neighbourhood watch, but it was from the bottom up and they had a 75% reduction in burglary over three years. But that was a very particular kind of neighbourhood watch and it required thought.

Q320 Jeremy Corbyn: In my area, the crime is incredibly localised. If you go 500 metres away from a high-crime area, there is no crime at all.

Professor Laycock: There is a lot of work on hotspots, which we do not have time to go through.

Q321 Mr Chope: Can I ask a separate but related matter relating to foreign national offenders? There are an increasing number of foreign national offenders in our prisons. What can we do in relation to crime reduction policies related to them? Do you think just the same principles apply to them, or do you think there should be a particular focus or a different type of crime reduction policy addressed to foreign national offenders?

Chair: That is a question that silences the panel.

Mr Chope: Pass it down the line.

Professor Laycock: I am not often speechless. I would want to unpick that assertion a bit and find out how many exactly are there, and where are they and where are they-

Q322 Mr Chope: A lot of them are in prison but a lot of them have not been detected yet.

Professor Laycock: If you talk to British Transport Police, for example, about crime on the underground, they will tell you that there are a lot of foreign nationals stealing from people on the underground. It is a big problem for them and they make every effort to deal with that, so they are aware of it. This is why the problem definition is so important. I don’t know of anyone who has actually looked very closely at it, but it is a very good issue to look at.

Chair: We are very grateful to the four of you for your help this morning. It has been really interesting and will feed into the work that we are now doing. Thank you very much indeed for giving us your time.

Prepared 27th January 2014