Examination of Witnesses (Question Numbers
647-686)
Professor Ken Pease OBE, Professor Hazel Kemshall
and Professor Carol Hedderman
8 June 2011
Chair: Good morning, Professor
Kemshall, Professor Hedderman and Professor Pease, and welcome.
We are grateful to you for coming in and giving us your experience,
knowledge and fruits of academic research in the area we are working
on, the work of the Probation Service. I will ask Claire Perry
to open the questions.
Q647 Claire Perry:
It is very nice to talk to some people who perhaps have stepped
back a little bit from the hurlyburly and the overloaded
case management system. We are very grateful to you for taking
the time to come and talk to us. What I would like to know, to
start with, is what, in your view, does the research show the
Probation Service is doing well and what are the main problems
and shortcomings that you see—if we can start at a sort
of over50,000 foot level initially?
Professor Kemshall:
Good morning. I have a few bullet points about what I think the
research shows probation does well. The first is the community
supervision of sex offenders, particularly through specialist
teams and units, group programmes for sexual offenders and, to
a limited extent, also violent offenders. The evidence also shows
that highrisk public protection teams, often comprising
both police and probation colocated together, have been
beneficial. The multiagency public protection panels for
multidisciplinary risk assessment and management have also
been helpful. When applied consistently, OASys has also proved
helpful in identifying risk, particularly risk of reconviction,
but more recently the risk of harm. Targeted work on what are
called the criminogenic need factors—although I prefer risk
factors—based on the effective practice research agenda,
has also had a good track record. Do you want me to talk about
the limitations?
Claire Perry: Yes, please.
Professor Kemshall:
The first one would be lack of consistency across probation trusts.
When you look at the performance measures, that would also be
borne out, particularly in applying that effective practice agenda.
I would also personally say that work, both one to one and in
terms of group programmes, with violent offenders has lagged behind
work with sexual offenders. That has often been because the policy
and media focus has been on sexual offenders, particularly over
the last 10 years, rather than necessarily violent offenders.
Claire Perry: Interesting.
Professor Kemshall:
There have occasionally been problems with attrition from treatment
or programme interventions, particularly where the waiting time
for the offender to join those programmes has been too long.
The third point would have to be about there still
being some inconsistency in the application of OASys, both in
risk assessment and in case management.
Q648 Claire Perry:
Thank you. Would you like to comment, Professor Pease?
Professor Pease:
Only as a limitation. If one believes the Ministry of Justice
statistics, it is the case that the Probation Service disposals
do not reduce reconviction rates relative to statistical expectation.
Chair: Could you speak
up a little and repeat what you just said because the door was
shutting?
Professor Pease:
I am sorry. I have one comment because my expertise is more on
the statistics of the thing than it is on the daytoday
operation, as Hazel has mentioned. If one believes Ministry of
Justice statistics on rates of reconviction, it appears that one
can predict probability of reconviction on the day of sentence
as well as one can at the end of sentence.
Q649 Claire Perry:
Effectively, there is no impact of the probation intervention?
Professor Pease:
Taken across the board, that is so, I think.
Q650 Claire Perry:
Interesting. Professor Hedderman, do you have something to add?
Professor Hedderman:
I would disagree with that in that, in this field, all you can
ever do is compare. You are never in the position of saying probation
versus nothing. You are always in the field of saying probation
versus either a short prison sentence or a fine. The Ministry
of Justice have recently published a report which looked at what
happens when you put likeforlike cases on a short
prison sentence or on probation—allowing for the fact that,
usually, you get different people getting different sentences.
That seems to show that the impact on reconviction of probation
and suspended sentence supervision is greater than a short prison
sentence.
Q651 Claire Perry:
We have just come from an interesting breakfast meeting about
short sentencing, but that is a whole other topic. Thank you.
I wanted to probe on one thing you said, Professor
Kemshall, about lack of consistency across Probation Services.
If you think of supermarket chains—and we have lots of Tescos—if
there is a failing Tesco, the management of Tesco works out why
it is failing and does something about it, or at least brings
it up to a group average. Why is that inconsistency of result
allowed to persist? Is it a failure of inspection or is it a failure
of these organisations not talking together or some other outside
factor, in your view?
Professor Kemshall:
I don't think it is a failure of inspections. We have a lot of
inspection reports and a lot of effective practice reports. Perhaps
it is a failure to implement the learning from inspections, first,
and perhaps also a failure to recognise that the Probation Service
operates through individual probation trusts rather than as a
single national unit in that sense. If you have that pattern of
delivery, then you will get variation. It is a speculation, but
I would suspect that the best performing areas in effective practice
have senior management leadership and drive about that particular
agenda.
Q652 Claire Perry:
Thank you. Could I turn to risk management systems? We know the
Probation Service has been very focused on risk reduction, and
your point about media hype—meaning the risk focus on sex
offenders was increased—is a very important one. Do you
think the existing probation systems, and in particular the tiering
of the staff structure, which is something we keep coming across,
have helped allocate the resources effectively? Related to that,
do you accept the idea of trying to push a bit more responsibility
and broadness of approach down into the Service is a good thing
to do?
Professor Kemshall:
Tiering is an inevitability. Resources must follow risk and resources
have to be rationed. There is no way round that. The resources
for probation are finite and, in fact, are shrinking. There is
always going to be a need to target resources most effectively
at risk. OASys has helped with that tiering system—the level
1, 2, 3 and 4 tiering system—but there are two problems.
The first is that OASys can still be inconsistently applied across
all the trusts, and even, indeed, within a trust. Although it
is still an important KPI for probation, one could see the use
of that KPI as driving that quality up, hopefully. Also, one might
wonder whether the tiering at levels 3 and 4 is exactly right.
There tends to be, on occasion, a confusion about high risk of
reoffending and high risk of harm. That can be a problem in how
that tiering works.
Q653 Claire Perry:
That leads into my final question. As we see potentially more
involvement of voluntary and private organisations in managing
the offender, do you think the duty of protection can be adequately
controlled in those transfers? Is it a question of making sure
OASys leads you to the right people to be passed over, if you
like? What would be your view on how that universe of providers
could work together to make sure the risk of harm is reduced or
minimised?
Professor Kemshall:
A decision would have to be made on the level of risk beyond which
you could not pass that case—that person—on. You might
decide that only low or medium risk cases were going to be passed
to the third sector to supervise. In some senses that already
happens. Probation trusts do commission services from a range
of providers and partners to do that.
Taking the question as a whole, I don't think we
can be sure, one way or the other, about the competence and capacity
of the third sector to deal adequately with levels of offender
risk. We do not necessarily know whether there is a competence
and skill gap there to do that. Commissioners of those services
would need to be assured of that through a rigorous tendering
process. If you are going to pass a risk you need to be sure that
the provider can adequately deal with it. That mechanism would
have to be there.
Q654 Jeremy Corbyn:
Welcome and thank you for coming to help us with our inquiries.
This question is to Professor Pease, but I would like to hear
comments from your colleagues. What evidence do you have for suggesting
that community sentences offer no measurable level of public protection?
Professor Pease:
There are three that I would like to compare in evidence. The
first is the Ministry of Justice analysis, which uses the Copas
statistical predictor of reconviction, which you will find up
to and including, I think, the 2009 adult offending cohort. There
you will find that statistically predicted and actual rates of
reconviction for the community sentence group are almost exactly
identical.
The second one, as Carol Hedderman rightly says,
shows a slight superiority of community sanction, or, rather,
probation, over short prison sentences. It also shows superiority,
of course, of long prison sentences over short prison sentences.
The problem with both those analyses, and indeed every other analysis
which has been done in this country, is that they cannot take
account of things that are not in the dataset which the court
takes account of. It is a statistical predictor, but if the court
takes account of relevant factors above and beyond the factors
which the statisticians take into account, there will be spurious
additional treatment effects. The effect that Carol Hedderman
refers to is very interesting because it is very age specific.
It is very, very small for young people and substantially larger
for older people.
The third kind of evidence that I would call attention
to is this. You will be aware of the Campbell collaboration—it
is like the Cochrane collaboration in medicine, whereby the methodology
of a variety of studies is compared and the gold standard, which
is randomised controlled trials, is used to reach conclusions.
The person who was commissioned to do the analysis of short custodial
sentences against community sanctions was under the direction
of Professor Martin Killias, a Swiss criminologist, and 300 studies
were reviewed, of which five met the standards of randomised controlled
trials and showed no statistically reliable difference in rates
of reconviction. There were some strange individual effects, but
I probably shouldn't go into those because it would take me some
time.
Those are the three. The Campbell collaboration result,
which is randomised controlled trial-based—from five studies
worldwide because it is very, very difficult, some would say unethical,
to make random allocation as between short custodial sentences
and community sentences, so there are only five worldwide—demonstrated
no reliable difference between the sanctions. I always feel I
need to make a caveat here. I don't want it to be like this. I
want community sanctions.
Claire Perry: Neither
do we.
Professor Pease:
Quite so.
Q655 Jeremy Corbyn:
Do you mean this is an inconvenient truth? Is that what you are
trying to say?
Professor Pease:
I believe it is a truth. Carol may disagree. I am convinced it
is a truth and it is one that I would wish to change by incentivising—I
am very interested in the social impact bond approach, for example,
and ways of incentivising community administrators of penalties
to make things better—but it feels like a truth to me.
Q656 Jeremy Corbyn:
If we accept your analysis, have you done any further research
on the quality of life of exoffenders later on between those
that did custodial sentences and those that did community sentences,
such as educational attainments, achievements, family, etcetera?
Professor Pease:
Yes. These were the perverse results from the Swiss study of
Martin Killias that I described, because they show a marginal,
nonsignificant superiority. I should make it clear that
this was a comparison, in this case, against community service,
not against probation. That comparison showed that short custodial
sentences had slightly more reconviction, but not statistically
reliably, for the first five years. However, for the following
six years the thing reversed, so that people who had had a short
spell in custody showed better social adjustment in years 6 to
11 than people who had been given community service. I do not
know how to interpret that.
Q657 Jeremy Corbyn:
How about your colleagues?
Professor Kemshall:
There are two things I would like to add. My area of interest
is the most harmful—the highest risk—offenders. It
is worth remembering that in 2009, for which we do have figures,
only 0.26% of the probation caseload out of some nearly 180,000
offenders who would have been supervised in the community at that
time, went on to reoffend seriously harmfully and to necessitate
what is called a serious further offence report.
Also, there may be other ways of thinking about this
issue. For example, I am a member of a probation trust board which
is either very close to the top, or indeed the top, in terms of
the performance measure on reconviction. The way in which that
is measured currently for that service is to contrast the predicted
rate taken from the risk assessments in OASys: in other words,
what it is expected those offenders will do and what, over each
quarter, they do do. This particular service performs very well
in terms of people doing less than is expected of them. It may
be quite interesting to contrast the very best performing trusts
on that measure against those which are at the bottom, and to
ask whether there is anything different, in terms of what they
do to achieve those rates, that could help us here.
Professor Hedderman:
I understand Ken's point about random controlled trials. The difficulty
is that that sort of approach comes from such things as trialling
medicines where you expect a heart drug to basically work the
same way with people. That is true in controlled trials for medicines,
but when you deliver it in the community you find that people
who are more overweight do more or less better on it, or that
older people may do more or less better on it, so it is a prescription
issue as well as an effectiveness issue. Your GP has to accurately
decide what sort of medicine is suitable for you.
What the random controlled trial thing does is to
say, "We are going to ignore the expertise," or, "We
are going to ignore what the sentencer thinks should happen in
this case. We are going to randomly give you either a custodial
sentence or a community one," and "Oh look, it's ineffective."
It doesn't seem to be more effective because you haven't taken
into account the particular circumstances of the individual. I
am not convinced that random controlled trials are the best way
of evaluating something like a patient. Aside from the ethical
issues, there is a question of: do you expect a probation or prison
sentence to work equally well for different people?
Q658 Jeremy Corbyn:
Breaking the Cycle, the Government's Green Paper, calls
for more effective and robust community sentencing. Do you have
any views on this and do you feel that that is going to make much
of a difference? That is to any or all of you.
Professor Hedderman:
Sentencing is intended to do a number of different things, and
every individual sentence is expected to do something, or probably
all of them. There is a distinction between whether that aim of
sentencing will have an impact on future behaviour. "Robust",
to me, tends to sound like it is a punishment. There is a difference
between holding people to account and being punitive. I think
holding people to account is quite an effective way of starting
to work with them to reduce their reoffending. Although I don't
have a problem with punishment in itself, it is quite important
to distinguish whether punishment is effective in reducing offending.
My own view is that it is not particularly effective. Certainly
short doses of prison do not strike me as effective.
Professor Pease:
Could you repeat the question? I beg your pardon.
Q659 Jeremy Corbyn: Yes.
In Breaking the Cycle the Government emphasises the need
for more effective and robust community sentences. My question
is: will this make much of a difference in your view? Do you think
it is a helpful suggestion from the Government?
Professor Pease:
During the 1970s I was head of the Home Office Research Unit's
social work section and therefore responsible for various treatment
initiatives. One was called IMPACT—intensive matched probation
and aftercare treatment. The thing that frustrated me then—this
plays into what Carol said—is that interactions are really
important. To give one example from that era, it was the case
that certain personalities of probation hostel warden were much
more effective at stopping absconding and reducing reconviction
among their charges than others. I asked repeatedly over the next
20 years whether these personality factors were ever used in selecting
probation hostel or other wardens and, uniformly, they were not.
If you like, the raw material for working out the interactions
that Professor Hedderman refers to simply were not followed up.
That is why the incentivisation of the Probation Service and other
community agencies to do this strikes me as really important.
Again, the social impact bond strikes me as an interesting way
of developing this point.
As to the ceiling of that effect, I really don't
know. Whether it will confer equal levels of public protection
relative to custodial sentences is perhaps open to question, but
it is certainly something worth striving for.
Professor Kemshall:
That is why it might be worth comparing the best with the worst,
because it might tell you what it is that people do that underpin
the community sentences that are making a difference.
Q660 Mr Llwyd:
Does the assessment of reconviction rates provide a reliable and
robust means of measuring the effectiveness of various sentences?
Professor Pease:
It seems to me that it is the only one the public has any right
to look at. One's lifestyle is one's lifestyle and reconviction
is when people intrude on the lives of others. My concern is that
it is not a level playing field in terms of the way in which it
is counted. Reconviction rate is counted from the date of release
from imprisonment and it is from the date of sentence for community
sanctions. That discounts the one thing that custody does do,
which is to keep people away from those on whom they predate for
a short while. It seems to me that a level playing field would
require counting from the date of sentence whatever the sentence
is. That would, of course, make the benchmark for community sanction
success much higher. That doesn't particularly worry me because
it is a more accurate statement of public protection afforded
by different forms of sentence.
Professor Hedderman:
Can I put a different alternative to you? If you want to have
a level playing field, the alternative way of doing it would be
to not count reconviction until the sentence ended. If you think
of probation as being like a course of antibiotics, it is unreasonable
to expect it, on day 1, to have its effect. The alternative would
be to wait until the probation order has ended and wait until
somebody comes out of custody. If you measured it that way, probation
would come out looking significantly more effective than a short
custodial sentence.
The simple fact about reconviction measures is that
it is the best one we have. It isn't in any way perfect. Basically,
you have a choice between accuracy and transparency. The more
accurate you want to make it, the more complicated the analyses
get about what exactly you are measuring with what. It is, at
the moment, the best thing we can do.
Q661 Chair:
Although there is a difficult decision to make about which measure
to use to apply payment by results, there is no reason, is there,
why we shouldn't have all three of these forms of information
available to us as a general means of assessing?
Professor Pease:
I agree, but I still like the "from the point of sentence".
This is only an argument by analogy, and therefore flawed, but
if you are looking at five-year survival rates for medical treatments,
one of which involves hospitalisation and the other does not,
then you start counting from the point of diagnosis, not from
the point of discharge from hospital, which would give an advantage
to the others.
Government statistics are vulnerable in three ways
in relation to crime and justice. This will be one sentence, I
beg your pardon. First of all, it understates the extent of crime,
for example, by discounting serious crimes in the British Crime
Survey. If you believed what people told you, which is what the
British Crime Survey is supposed to do, then those who are chronically
victimised would be much greater—a woman who has been hit
by her husband every Friday night for the last year will have
suffered 52 times, but the British Crime Survey will count it
as her having suffered five times, so that is truncated. If you
believed people, that would be much greater. The second is it
understates the degree of inequality between areas and people
in the extent to which they are victimised. The third, because
of its use of, for example, prevalence measures—reconvicted
or not reconvicted—rather than the number of offences for
which they are reconvicted and those which are taken into consideration
and others—Yes, Sir Alan?
Chair: I am having difficulty
hearing you.
Professor Pease:
Yes, okay. In all those three cases, it seems to me that official
statistics understate the extent and the inequality of distribution
and overstate the effectiveness of penal sanctions.
Ben Gummer: I missed that
last point as well.
Professor Pease:
I am sorry, I was trying to be brief.
Chair: The third of the
three.
Professor Pease:
The third one is the effectiveness of criminal justice systems.
If one looks back at fine payment rates, to take one example which
is not too contentious, I think, it is one which was savaged by
the Public Accounts Committee and the National Audit Office a
couple of years ago to the point that we do not know what proportion
of fines imposed are paid and, if so, in what rate; there is also
a set of perverse incentives to administratively cancel fines
in certain circumstances. I use that as one easy example because
your fellow Committee has made that point hitherto. I could produce
other parallels in terms of things that are ignored. For example—
Chair: That is a long
enough diversion from Mr Llwyd's line of questioning, so I will
put him back in charge.
Professor Pease:
The reason why I went in that direction—I apologise that
it was a diversion—is to say that reconviction is one figure
but if you say "reconvicted/not reconvicted" you are
understating it, as is typically the case in crime and justice
statistics relative to the amount of crime which people suffer.
Q662 Mr Llwyd:
But there is no unanimity as to when the period should start.
You take your view and presumably another view is taken by other
members of the panel. Your view is that it should start from the
point of sentencing. You have explained that position.
Professor Hedderman, you have been critical in the
past of the use of crosssectional snapshot samples used
for measuring reoffending rates rather than the conventional longitudinal
approach. Can you explain what the difference is and why the choice
of technique is so important?
Professor Hedderman:
The reason most studies take commencement is that we know most
people who reoffend after a court order or release from a prison
sentence do so within about the first six months and the majority
of their offending will certainly happen in the first year. If
you take crosssectional samples from the probation caseload,
you are taking people who were sentenced yesterday, who are at
very high risk, and people who have been on caseload for three
years, whose risk is quite low, so it undercounts the extent to
which the Probation Service are getting fresh cases in and dealing
with high risk. That is the first thing.
The second thing is that the Probation Service tends
to weight its work towards the front end for that reason. Even
if they are getting credit for the people who have been there
for three years they are not doing, usually—unless they
are very serious cases—a lot of intensive work with those
people. They have done that in the first few months. It is falsely
suppressing how much reoffending those people are doing. It is
this transparency thing as well, in that it gives you yet another
set of different figures that look much lower than the annual
reconviction rate for probation. The study I did in the East Midlands,
when I looked at the overall East Midlands reconviction rate using
a standard approach, came up with pretty much the national level
of reconviction. I feel it is easier for people to understand
if they see that the national rate is about 38% and the local
figure is about 38%. They can see that they are coming from the
same sources.
Q663 Mr Llwyd:
Your view, presumably, will be important in terms of modelling
payment by results.
Professor Hedderman:
The payment by results approach will be done on a commencement
sample, yes, not a caseload sample.
Q664 Mr Llwyd:
I beg your pardon?
Professor Hedderman:
The Peterborough analysis will be done on a longitudinal approach.
Q665 Mr Llwyd:
On a longitudinal basis. That is presumably what you would like
to see rolled out generally on payment by results.
Professor Hedderman:
Yes. It is only the local reconviction comparison. The area conviction
is the only one that takes a caseload approach. All the others
take the standard approach.
Mr Llwyd: Thank you.
Q666 Elizabeth Truss:
I want to ask a question about payment by results. It sounds,
from what you have been saying, Professor Pease, as though the
only effective interventions are where it does not only involve
the Ministry of Justice but wider public agencies; for example,
the DWP and the Work programme or, perhaps, social services or
the NHS. How could a payment by results model work across those
services as well as just the Ministry of Justice?
Professor Pease:
I don't think I am at all equipped to answer that question—not
remotely—because the community safety partnerships, which
I am more familiar with, have enormous difficulty, and the Department
of Health is also extremely relevant in such things. The point
is that payment by results would at least give an incentive for
change, which I believe is now limited to the professionals—the
community service and probation administrator.
Q667 Elizabeth Truss:
You also have payment by results working in other areas. For example,
somebody who has been convicted of a minor offence may also be
part of a payment by results programme to get back into work.
Could those things be integrated? Would that be helpful?
Professor Pease:
With the motivation, I see no reason why not. There are two caveats,
if I may, and one is an historical one—the Californian probation
subsidy scheme, which should be looked at by the Committee in
written evidence. It tried payment by results of a kind in 1965
that fell apart in 1970, for reasons that remain in dispute but
include some of the difficulties that you half allude to.
The second one is that, when it is evaluated, it
seems to me important that you get some grizzled practitioners
to look at it because there are all kinds of ways and means by
which you can distort the figures by early terminations and back-loading
stuff—all kinds of things. You need a grizzled practitioner.
I fear I have not answered your question.
Q668 Elizabeth Truss:
On the subject of community sentences versus custodial sentences,
you have said that community sentences are not effective in reducing
reoffending, essentially.
Professor Pease:
Not relative to other things, as Carol Hedderman correctly says.
Q669 Elizabeth Truss:
What about the cost benefit of them? Clearly, the costs are lower.
If you had to do it on a costbenefit basis for a particular
person being convicted of a particular crime, how would you assess
one against the other?
Professor Pease:
I did a piece of work for Civitas about six months ago which
tried to calculate that and which I would be happy to supply the
Committee with. What it showed, effectively, was that it depends
on how many undetected crimes people have committed. The Home
Office has costings of crime, and if one applies those costings
of crime alongside the sanctions, then what I got it at was something
like, on average, five and a half to six undetected crimes. If
somebody, for every conviction, commits more than, say, six undetected
or unconvicted crimes, then the costbenefit analysis comes
down in favour of custody.
Q670 Elizabeth Truss:
Could the rate of detecting those undetected crimes improve by
the use of more halfway house solutions, for example, tagging
and you mentioned probation hostels earlier? What about those
types of solutions where people who have been convicted aren't
necessarily in custodial sentences but are being more closely
monitored? Have you investigated the effects of those kinds of
interventions?
Professor Pease:
I have not. Does anybody know? Can the collaboration stuff help?
Professor Hedderman:
There is an evaluation by an organisation called Matrix, which
comes up with slightly different results to Ken. It basically
says that community supervision, particularly with a form of monitoring,
is more cost beneficial than prison.
Q671 Elizabeth Truss:
Do you have that information?
Professor Hedderman:
I can get that sent to you. You have got it.
Professor Pease:
I would make one comment on that. That is true and the methodology
is excellent. The trouble is that the standard for community sentence
reconviction rates is not taken from Ministry of Justice statistics.
It is from Campbell collaboration best performance statistics,
so it is an unrealistically optimistic statement of the effects
of community sanctions in that paper.
Q672 Elizabeth Truss:
Do the panellists think it would be helpful if courts, when putting
together sentences, had the information about the cost benefits
of the sentencing for the particular crimes committed? Do you
think more research should be done into the area to understand
what the costs and benefits are?
Professor Hedderman:
My answer is an unequivocal yes. The courts are not particularly
well informed about the effectiveness of different penalties.
When you ask sentencers what they would like, they don't really
want reconviction rates. They want to know about the cases they
themselves have sentenced. It is quite a difficult thing to persuade
them that it is information they would find useful. I don't think
they see that need themselves at the moment.
Q673 Elizabeth Truss:
Presumably, they would want to know the efficacy of the sentences
they are passing.
Professor Hedderman:
I think not, actually. I have been interviewing judges and magistrates
over the last summer and their view is that having passed sentence—and
this is particularly true of magistrates—they have done
their job. What happens thereafter is the responsibility of the
Probation and Prison Services. Different magistrates might have
a different view, but that was the consistent view I got from
the interviews I conducted.
Professor Pease:
Yes. I support that. In the old days—and it may still exist—there
was a little booklet which was given to all magistrates called
"The Sentence of the Court" and it had an annex by Dr
W H Hammond about the reconviction rates after different sentences.
Whenever I talked to magistrates, I always said, "Have you
read the annex in 'The Sentence of the Court'?" I have never
found one hand go up in any magistrates meeting that I ever attended
to say that they had read the statistics at the back of the book.
Q674 Chair:
The community courts work on a different principle where that
kind of experiment has been operated, as in Liverpool. You have
members of the judiciary who see it as part of their job to follow
up the offender through various stages and establish whether they
are making progress or whether the sentence has to be varied.
Professor Pease:
That is Carol's point, isn't it, that they are only interested
in their own outcomes?
Professor Hedderman:
The difference with a community court is that their approach is
problem solving. Most courts are there to punish people and for
public protection. That is not quite the same thing as a problemsolving
approach where they are focused very much on the individual in
front of them and what might be done to change that person's behaviour.
Q675 Chair:
But don't magistrates say, "I don't want to see you in this
court again"? I'm quite surprised that you say that.
Professor Pease:
That is to the person in front of them, not everybody else. That
is an indication to the individual.
Q676 Elizabeth Truss:
I want to ask a final question about the international evidence
on this and also the way international systems work. It seems
to me, with a cursory look at it, that some other countries have
more of a joint correctional service which will have a combination
of custodial and noncustodial sentences, so more of a continuum.
Do you think a system like that would be better in Britain rather
than the very separate sort of custody and probation regimes we
have?
Professor Hedderman:
We all thought that is what NOMS was going to achieve and has
not come anywhere close to achieving.
Q677 Elizabeth Truss:
Why hasn't it? Do you think it is a good objective?
Professor Hedderman:
There is an interesting phenomenon in Britain: you take the results
of research, the Government take them on and they somehow reinvent
them in their own image. I don't think anyone would disagree with
the idea that you should have seamless management, but somehow
that becomes an organisational and bureaucratic objective. The
same is true of offender management. The idea behind offender
management is that it is better to have one person dealing with
the offender. They build rapport and relationships. What you get
as an offender management is the bureaucrat who is in charge of
linking all the different little bits together. The concept never
seems to get turned into practice; it gets bureaucratised. There
are a lot of lessons we could learn from abroad and that seamless
approach—
Q678 Elizabeth Truss:
Which country would you pick from a hat?
Professor Hedderman:
I think Australia has probably more of an understanding of the
relationship that you need to build with the offender. On the
organisational front, I am not sure I am convinced that structures
are necessary. Our Probation Service and our prisons do communicate
pretty well with each other at an individual level but, because
it is at an individual level, it may not be consistent. That is
Hazel's point, that there is inconsistency.
Professor Kemshall:
It also has to be said that here, at home, it is interesting
that the police and probation have made quite a strong partnership,
particularly around integrated offender management—which
I think you will hear about later this morning—and also
in terms of public protection. It shows that MOJ agencies can
work together and it is interesting perhaps to have some lessons
about how those two do.
Q679 Elizabeth Truss:
Can I get your views on the international point?
Professor Pease:
I have nothing helpful to add.
Q680 Chris Evans:
I want to focus on training and I direct my question to you, Professor
Kemshall. The new probation qualification award was introduced
in April 2010. What is your view of how it is working? Are staff
being given significant time to train and is there significant
supervision and support?
Professor Kemshall:
It is a little too early to tell how well it is working. It came
on stream in 2010 and we need time for the cohorts to go through,
to evaluate that properly and to look at whether the structure
is going to work particularly well. It does bring some very particular
benefits here, especially to probation support officers, PSOs,
because, perhaps for the first time, it is giving them a coherent
structured training pathway, particularly up to level 3. It gives
a clear pathway of both training and career progression for that
grade to probation officer level. It is also quite important for
PSOs in that it relates to the tiering question asked by Mrs Perry.
PSOs are now increasingly taking on a lot of community supervision
and engaging with high risk cases and we know, from previous inspection
reports, that that is a grade that needs to be trained well in
terms of risk.
Q681 Chris Evans:
I was concerned that the 2005 Napo survey stated that 72% of
responders said there were too few opportunities for staff development.
This was echoed later by research at De Montfort University. Why
do you think this came about—
Professor Kemshall:
Could you repeat that, please? I am struggling to hear you.
Chris Evans: I am sorry.
Perhaps it is my accent, is it?
Professor Kemshall:
No. I am struggling to hear you.
Chair: It's the acoustics
in this room, I think.
Chris Evans: It is the
acoustics. I will speak up. I am often accused of shouting, so
I kept my voice down.
Napo's survey of 2005 said that 72% of respondents
felt there were too few opportunities for staff development. This
was echoed later by a De Montfort University study. Why do you
think that situation came about, or why do you think there was
that belief there?
Professor Kemshall:
Do you mean in terms of staff development for people in the workforce
currently?
Q682 Chris Evans:
Yes, why did they feel there were not enough development opportunities
there for them?
Professor Kemshall:
A proportion of that is going to relate to the issue I have just
spoken to, which was the grade below probation officers, PSOs,
the Probation Service officers. They are slightly confusing terms,
I think, which is a little unhelpful. Traditionally, people had
to leave that grade and leave their employment to go and train.
There was no progression beyond that grade. You could move within
it and receive incremental payments and so on, but there was no
progression through that grade. Also, in terms of staff development
within grade, I would suspect there are difficulties, in terms
of inservice training, of people feeling that there are
developments within two particular specialisms; that people are
building a portfolio of different skills through their work career.
Q683 Chris Evans:
Do you think probation officers have enough training? Do you think
it is about right now or do you think this is an area that should
be looked at again?
Professor Kemshall:
Did you ask me about probation officers?
Chris Evans: Yes. Do you
think they have enough training or do you think this is an area
where there should be a system of continuing professional development
or something like that?
Professor Kemshall:
Yes. We don't know yet how well the PQF is going to work, for
the reasons I explained earlier. Personally, I think it would
help if we could see a clear pathway of building competence between
years one and five within the workforce. We should not accept
that, after the training period, I enter on day one and I am equipped
to do everything. That is clearly not the case. I may need to
learn a range of skills and competencies to deal with more challenging
types of offender and offence types and more challenging types
of risk throughout my career. At the moment, we don't see that
pathway and an awful lot of inservice training is by selfselection—members
of staff choosing what they want to do—and much of that
training is not then assessed in terms of competence on the job.
We train but we don't know the impact of that inservice
training well enough.
Q684 Chris Evans:
We have heard in the past, from other people in front of us, that
there seem to be people recruited into the Probation Service but
a high level of exit out of it. Why do you think that is? Do you
think there is a general perception of probation and that there
is a gap between what probation officers are prepared to do and
what people think probation officers do? If so, how do we overcome
that perception?
Professor Kemshall:
Can you repeat the first part of the question, please?
Chris Evans: I am sorry,
the table is creaking as well. I'm having all sorts of problems
here. We have been told by people in the past here that people
enter the Probation Service but there is high level of exit through
being disillusioned, demotivated or whatever. The general feeling
that has come out is that people have a perception of what a probation
officer should do compared with what a probation officer actually
does. How do we address that fundamental issue of perception?
How do we get people to understand what is expected of them when
they enter the Probation Service?
Professor Kemshall:
Recruit better.
Q685 Chris Evans:
How do we recruit better? What types of people are we looking
at? What areas do we recruit from in particular? What I am asking,
essentially, is this. If you had somebody in front of you as a
probation officer, what qualities would you be looking for?
Professor Kemshall:
To people who came, I would ask the question, "Why do you
want to join the Probation Service?" If their first answer
to me was, "Because I want to help people," or, "I
like people," or, "I wish to work with people,"
that would worry me greatly. I would expect people to be able
to express some understanding about the role and responsibility
that they are taking on. It is a very big responsibility.
I am very struck, for example, when I train probation
trainees and we are talking about risk and start to look at the
case studies, that people can be extremely surprised about the
sorts of people and the sorts of things they have done that, in
a very few months' time, they may be being asked to deal with.
I think there is a misunderstanding about the challenges faced
within the job. You need high resilience because there can be
a lot of failure. Sadly, people do come back. People don't always
do what you think is right for them and, for some people, change
is a very long journey; but also some people have done some terrifically
awful things that have to be faced, talked about and changed.
Q686 Chris Evans:
Have you done any studies into why people exit the Probation Service?
Professor Kemshall:
Not personally. De Montfort has done those—I can supply
them—and, indeed, they have been done by other researchers.
There are issues about morale and about personal resilience to
the challenges of the job, and particularly this issue of disappointment
and failure. I think there are misperceptions about the workload
and what that is going to look like and misperceptions about what
some perceive as a mismatch between the facetoface
contact work and the administration and bureaucracy of the job,
which is inevitable.
Chris Evans: That would
be interesting to have.
Professor Kemshall:
Yes, of course.
Chair: We need to move
on.
Chris Evans: Thank you.
Chair: We are all very
grateful to the three of you this morning. We have had to compress
things in a short time but we are very glad of your help. Thank
you.
| 
