How 'transparent' are personal budgets?

[Note: The research referred to in this post has been expanded upon since it was written and published in the Journal of Social Welfare and Family Law (Series, L. & Clements, L. (2013) 'Putting the Cart before the Horse: Resource Allocation Systems and Community Care', Journal of Social Welfare and Family Law, 35(2) 207-226).  A pre-print version of this paper can be downloaded from here]

The personalisation agenda in adult social care is a bit of a hobby of mine.  It is massively important for understanding the landscape of adult social care today.  Sometimes, when I'm bored of the Mental Capacity Act, I indulge myself by doing little research forays into personal budgets, resource allocation systems, brokerage, providers and rules and regulations (see, e.g., this post on whether personalisation is inadvertently criminalising carers).  I know, I'm a very sad person.

The thing that interests me about personalisation is that it's simultaneously such a brilliant and yet dastardly idea.  The central idea is that instead of the local authority deciding what your needs are and matching them with community care services they have chosen, under 'self directed support' or personalisation you participate in assessing your own needs, then the local authority convert this into a cash value using a Resource Allocation System (a RAS), and then you spend that money on the services you think will meet your needs.  For people who would prefer to minimise contact with social services, who would prefer to research and choose their own services, who would like to be able to threaten service providers with taking their custom elsewhere instead of being stuck with a slot in a 'block contracted' service, for people who want to go to the gym or a football match or the pub instead of a day centre, for people who would like to employ a personal assistant directly instead of dealing with the vagaries of domiciliary care agencies - it is fantastic.  Well, almost.  When people have the knowledge, support and adequate resources to meet their needs and achieve their goals, personalisation is fantastic.

Personalisation is not without its critics.  And no, David Brindle of the Guardian, it's not just paranoid social workers who think it's a challenge to their role, and it's not just the Daily Mail fantasising about spending social care money on prostitutes and lap dancing  - there are concerns about personalisation from a range of perspectives.  The organisation In Control were the pioneers of personalisation, and in 2009 they held a conference entitled 'Don't be Fooled by the Law' which explored tensions between personalisation (which is a policy, not a law) and community care law.  Concerns were voiced from a range of legal professionals around whether pure self-assessment would miss important issues, whether personal budgets attracted safeguarding issues, who was liable if a person was placed at risk, and also whether eligible needs would still be met under personalised models of care.  The issue I want to focus on today, however, is whether personalisation is as 'transparent' as it's advocates suggest, and whether it does offer opportunities for 'cuts by the back door' as its detractors say.

Transparency is central to the ideals of personalisation and may be legally important

[Edit: See Postscript 2 below for more information and a response to the issues raised in this post from In Control]

In their conference reported (Don't be Fooled by the Law) In Control stated:

In Control’s view is that FACS is not in fact “fair” and should be abolished, and that a RAS should become the single, clear and transparent way of determining the indicative resource allocation for individuals. 

They later addressed the concern that RASs are not transparent:

Q5. In reality the RAS is rarely transparent. Sometimes there are multiple versions of the Self-Assessment completed (“user,” “carer,” “care manager”), and then all the power is in the hands of the Local Authority to interpret this data and allocate resources as it wishes. 

Of course, any system is open to abuse and in a situation of budgetary pressure it would be remarkable if there were not examples of Local Authorities which found ways to make use of the reformed system to achieve short-term savings in ways which conflict with the Disability Legislation. Sometimes, individual managers who are under pressure to deliver savings targets do this, on other occasions there is little doubt that SDS is seen from on-high as a way to contain costs, and little more than this. This is extremely sad. In Control is absolutely adamant that it is unacceptable, and we are equally clear that it is not the Government’s intention in its Putting People First policy. Where we encounter such bad practice our approach is firstly to remind the Authority concerned of the Statement of Ethical Values which they agree to on joining In Control -this is the bedrock of all our work - and secondly to offer advice and guidance as part of our leadership development programme, and thirdly to advise them that they are liable to challenge under the Disability Discrimination Act 2005. If a Local Authority is challenged under this legislation, it may well be faced with Judicial Review Proceedings which will cost large sums of public money.

RAS's are touted as being more 'objective' than resource allocation done through traditional means; on a positive note this may mean that service users who are widely offered less attractive packages of support (particularly older people) may benefit.  However, it is in the nature of the RAS that some people will not benefit, because the resources are not being increased overall - they are being redistributed.  This is really clear from In Control's own chart from their guidance on creating a RAS:

The smoothed purple area reflects the new entitlements for service users calculated using the RAS, and the blue lines represent existing amounts spent on a traditional care plan.  You can see that some people have more money to purchase services, but you can see that other people's budgets have been drastically cut.  The interesting questions for lawyers is whether their eligible needs can still be met from this amount.  In the case JL v Islington (2009) the parents of a disabled child challenged a drastic reduction in his care budget following application of a sort of banded award system similar to RAS's.  The judge stated that the amount awarded at must be flexible, and if the amount arrived at by resource allocation procedures was insufficient to meet eligible needs, then it must be increased.  Imagine the cost to the system if all the people in who had their care packages cut successfully challenged it, meanwhile the people at the lower end were being awarded more...

In another important case, Savva v Kensington and Chelsea (2010), the judge stated:

The personal budget must be sufficient to purchase the services and is needs-led, and it seems to me that the only way in which a service user can be satisfied that the personal budget has been correctly assessed by the Panel is by a reasoned decision letter. [44]

and

That principle is directly applicable in the present context, and the problem of course, in this case, is that a reading of the letter from the authority, which states that the Panel has kept the allocation at the same personal budget, provides no indication whatsoever as to why the Panel arrived at this decision. [46]

The question is, if a local authority is relying on a RAS that is basically a mathematical algorithm based on multiple choice assessment responses, how 'reasoned' can this letter be?  A case that is headed to the Supreme Court will soon test this issue.  In KM v Cambridgeshire County Council the Court of Appeal described the process for arriving at an award using a RAS as 'tortuous' [2], [7], but did conclude that the council had sent an 'intelligible eventual explanation' and the decision to award that amount was rational.  The Lords of Appeal commented:

There has of course to be a rational link between the needs and the assessed direct payments, but, in our judgment, there does not need to be a finite absolute mathematical link. [22]

The Supreme Court will consider this issue further.  It will be interesting to see how far local authorities must make clear the reasoning involved in the RAS to service users to explain their awards, and whether this in itself is deemed a sufficient explanation.  It seems to me that unless the underpinnings of RAS's are made transparent, there is a real danger that the reasoning behind an award would become "computer says so".  In any case, however, it strikes me as a significant public interest issue to understand the general basis upon which community care resources are being distributed.

Why should we worry about what's in a RAS?

I would not for one minute deny that traditional systems for allocating resources in community care are highly subjective, and rife with opportunities for discrimination and unfairness.  However, I would contest the idea that RAS's themselves are either fair or objective.  They encode in their calculations certain assumptions about how much it would cost to meet a person's eligible needs.  This might include, for instance, assumptions about the cost of care services in the area, which may not take into account the cost of specialised services for meeting a person's needs, or it may inhibit purchasing from smaller or higher quality care providers that are more costly.  In a paper by one of the architects of personalisation Simon Duffy (2005) gives an example of a very primitive RAS in two tables on pages 7 and 8.  Note that people who live outside the family home get twice as much money as those who do not.  Now, clearly this is only a sketch and not a real functioning RAS, but it has clearly built in assumptions about the level of support a person will get from their family, with no consideration of how much support their family is able and willing to provide.  Some RAS's include a 'cost abatement multiplier' - which Luke Clements, in his excellent and entertaining paper on personalisation, describes thus: 

It is at this stage that any science in the process is jettisoned in favour of witchcraft. The resultant sum is then reduced. The reason for this has variously been explained as providing ‘headroom to avoid overspending’ or to avoid inducing ‘dependency’ (see L Clements, ‘Individual Budgets and irrational exuberance’ (2008) 11 CCLR 413–430). (p4)

It would be interesting to see what 'rational' explanations for cost abatement multipliers could be offered in court...

There are other reasons why it would be useful to look at RAS's.  Some people may need more help managing their personal budget, for instance through buying in brokerage services - it would be useful to know if this is factored into the award they are given.  If it is not, then we should seriously consider whether RAS's embed discriminatory practices against people who lack the mental capacity to manage their own individual budget.  It would be interesting to know whether a local authority operates the same, or different, RAS's for people who, e.g., are older, have mental health problems, have learning disabilities or physical disabilities.  It would be useful to know what kinds of questions in the assessment are linked to the RAS - are there salient issues that might affect somebody's community care needs that elude this form of 'multiple choice' assessment-by-numbers?  For all these reasons and more, I contend that it is very important we are able to explore the underpinnings of the RAS's operated by the local authorities.  But here's the problem.

[Edit: 02/03/2012 Since writing this piece I have continued to try to find out about RAS's for a research paper which I hope to publish at some point in the future.  My findings suggest that the these concerns about unfairness in the RAS may be well founded, but perhaps of more interest is the finding that many local authorities have found them to be so inaccurate they have simply abandoned using them.  Of those that continue to use them and supplied the relevant information, the gap between 'indicative amount' and 'actual allocation' is so significant, and so often people receive less than the RAS indicates, that the idea the RAS creates a 'fair and transparent entitlement' is laughable.  I wonder what the point of the RAS is at all, if care managers end up using different means to work out the cost of a budget?]

How transparent are local authorities' RASs?

As I said, personalisation is a sad little hobby of mine, so I wrote to various of the 'Total Transformation' local authorities who In Control worked with to test out the personalisation of adult social care.  I asked them various questions about how personalisation was going for them, and also for the mechanisms underpinning their RAS.  I didn't write to them all as this isn't a full blown research project and I didn't have the time, and I also wrote to Cornwall Council, who are not a transformation local authority, but it was where my doctoral research is based.  I made all my requests through the third party website What Do They Know, so they are available for all to view.  And yesterday I sat down and belatedly collated all the links and responses into this google spreadsheet.  The spreadsheet is (if I do say so myself) quite interesting for reasons unrelated to RASs, as it contains links to different authorities' policies around mental capacity and CRB checks, and also information on the number of legal challenges they have received in relation to personalisation.  Interestingly, Lancashire apparently received 30 legal challenges (including letters before action that didn't go to trial) last year, while Hartlepool received none, and [edit 07/09: I just spoke to Kent's FOI officer by telephone, who says they received a request this week asking whether they have had any legal challenges relating to the amount awarded in a direct payment.  She says that their legal team advised her they had had none to date.  Not even letters before action.]  Most areas seemed to be having around 5 or 6 personalisation related legal challenges a year; I'm not really equipped to say if that's a lot in the great scheme of things, but some readers may know.

The thing that is striking is that whilst most local authorities were happy to share policies and documents galore (sometimes nothing to do with what I had requested!), there was a marked reluctance to share information about the RAS in any great detail.  West Sussex and Hampshire Council were the most 'transparent', in that they provided information on the points that would be scored by different responses to the assessment questionnaire.  I have to say, I am still not clear from these responses how you get to a cash entitlement, it would have been nice to see a spreadsheet or something you could 'test drive', but certainly it is interesting to be able to see what questions are asked, and how much weight is attached to different responses.  Kent County Council responded that they did not operate a RAS.  Cambridgeshire, Hartlepool and Lancashire gave sort of vague descriptions of how they had developed their RAS's, but little detail on the content of the RAS or how it worked.  None of these responded to follow up requests for further information (although - if they received the requests - they would be required by the Freedom of Information Act 2000 to either treat these as new requests - that should have been responded to months ago - or as internal reviews meriting a response).  Interestingly, Cornwall County Council and Essex County Council applied exemptions contained within s36(2) Freedom of Information Act 2000 to supplying this information.  Essex have yet to supply the 'public interest test' itself, but I've chivvied them with a phone call.  Cornwall Council said this:

The RAS is a spreadsheet (at present) that uses questions in ACS overview assessment to produce a financial indication of how much it would cost to meet a person’s needs. If this was made available on the website then anyone completing the assessment may be able to answer the questions on the assessment in such a way as to produce a higher personal budget than they need. This could have serious implications on ACS’s management of their budget. This would also mean that a person’s assessment of their needs (our legal assessment tool to comply with Community Care Act) would not be accurate. The Council has considered the request very carefully and has decided that the information requested would, if disclosed, in the opinion of the qualified person, be prejudicial to the effective conduct of public affairs under S.36 of the Freedom of Information Act 2000.

This is quite an interesting response.  Basically, it seems as if Cornwall (and perhaps others) are concerned that by disclosing how their RAS calculates entitlements, service users will 'game' their responses to assessments in order to get their hands on more cash.  This strikes me as going entirely against the ethos of Personalisation, which emphasises trust in service users to determine their own needs and spend resources wisely in order to meet them.  It also strikes me that the public interest test has failed to take into account the public interest arguments I have outlined above: namely, that it is important that we are able to scrutinise the assumptions built into the RAS and understand how adult social care teams are distributing their resources.

[Edit 08/09: Update - Lancashire County Council responded to my request for more information on the mathematical formula used by the RAS to determine the indciative amount today by saying:

After having done an extensive search, I can confirm that Lancashire County Council do not hold any further information relating to your enquiry.

They state in their documentation that they do use a RAS, but they appear to be saying that they don't actually have any information on how it works. This seems absolutely incredible, so I've written a request asking them to clarify whether this is their position, or whether they're just applying an FOIA exemption and not sharing it with me.  I wonder if this is linked to their 30 legal  challenges last year!]

[Edit: 02/03/12: Cornwall Council contacted me to say that they had received a new request and decided to disclose their RAS after all.  You can find it on their FOI disclosure log.

Since writing this piece I have contacted 20 more local authorities, all of whom supplied details of their RAS, except Croydon.  Croydon did not say which part of the FOIA they were relying on to refuse to supply this information, but I will quote from their explanation of why:

Upon request we provide customers with quite precise details of how our RAS tool has calculated their indicative allocation. This of course is always specific to their circumstances and subject to a full explanation of the purpose and role of an indicative allocation. However, there are a number of reasons why it would be inappropriate for us to share or publish our precise formulas in their totality. The main ones are:

1.) The function of the RAS tool is to generate an indicative allocation to start support planning. Our RAS policy (which you have been sent) provides a far clearer and fuller explanation of our resource allocation system.

2.) If people completing the form (or advocating) were aware of how the answers impacted on the indicative allocation, then they may be incentivised to provide inappropriate answers to try and get a higher indicative budget. This could have serious implications for our assessment and support planning processes; and potentially even threaten the sustainability of our adult social care budget.

3.) The work of adult social care is in achieving personalised outcomes for individuals which are sustainable. The best way to promote this overriding objectiveis to focus on the community care assessment in the first instance. Over-emphasis on the RAS tool is very unhelpful from an operational and financial perspective (both generally and in individual cases).

4.) The RAS tool is developed and maintained iteratively based on learning from practice. Changes to the RAS tool can be made at any time it is felt appropriate. It is the RAS Policy that represents Council Policy, not the tool itself. The aim of these changes is to improve the reliability of the indicative allocations so that they can better inform support planning, etc. It is important to remember here that the RAS tool does not “allocate” anything. The production of an appropriate, outcome focussed support plan is the mechanism / system by which resources are allocated.”

Oh Croydon... I even wrote in my FOI request that it was for the purposes of research into transparency!  It's interesting that Croydon's refusal to disclose details on the RAS attaches to the ideological incoherence at the heart of these systems.  On the one, they are meant to create 'entitlements', and Croydon are worried that if people understood the RAS they'd rig their responses so they would be 'entitled' to more and the entire resource distribution system would collapse.  On the other hand, Croydon are quick to point out that the RAS doesn't, in fact, create a fixed entitlement.   "It is important to remember here that the RAS tool does not “allocate” anything", and so - the implication is - I don't really need to know about it anyway.  Thank you Croydon for such a fascinating response for my research on RAS's and transparency.]  One has to wonder, if the RAS "doesn't allocate anything" what the point of the costly and time consuming development and use of such systems is?

I have made many, many Freedom of Information requests in my time, and I have never felt as if I have hit such a brick wall in obtaining information as trying to find out more about RAS's.  This seems, to me, to be entirely out of keeping with the spirit of personalisation, and in a wider sense the transparency agenda promoted by this government.  Because I wasn't quick enough in chasing up local authorities for their responses, I didn't clock until last week that many of them had ignored my requests for clarification.  Unfortunately, this leaves me with a very short timescale as the clock for referring the matter to the Information Commissioner's Office will expire on 18th September.  Hopefully something more will come back by then, but so far the answer to my question seems to be: No, RAS's are categorically not a clear and transparent way of distributing resources.  Not yet, anyhow.

[Edit: 30/9: Essex County Council have now completed their 'public interest test' and have decided that it weighs in favour of disclosure after all.  Their RAS points scoring system has duly been disclosed here (pdf).    I'm still not sure how this relates to cash amounts, but I've sent a follow up request asking them.  After further correspondence, Lancashire have disclosed the full RAS system.  You can see their points scoring system here, and how points translate to indicative amounts here.]

Postscript

After posting this piece, I was interested to see these comments from social work students and practitioners on twitter (@thesmallplaces) about RAS's:

These comments suggest quite a concerning picture: that local authorities may be chopping and changing between using RAS's depending on what produces the lowest cost care plan; that practitioners themselves do not understand how the awards are arrived at; and that some RAS's may indeed reduce the awards available to people living with carers (although, these may be sensitive to what support carers are willing and able to offer).  In any case, there have been lots of reports on people's experiences of using personalisation - I'd like to see more on how RAS's are being developed and used in practice as well.

Postscript 2: In Control and transparency

Since writing this post In Control got in touch through twitter and drew my attention to their evaluation report for the transition project they ran. The relevant pages on RAS's are on p36-46.  Passages of interest include:

'Self-directed support will not function without a robust system that can put money in the hands of ordinary citizens in ways that are fair, transparent and efficient.' p26

'The RAS, whatever the finer detail, must contain at its centre a simple set of transparent rules to demonstrate how the individual gets a fair budget, according to their needs and social circumstances.'  p38

This extract from a chart on page 44 may also be of interest:

The report is really interesting in terms of exploring how RAS's are developed, and if you're a fellow RAS-geek it's well worth a read.

And for the avoidance of doubt, I asked In Control on twitter whether they thought local authorities should disclose the rules and formulae underpinning the RAS, and this was their response:

I should say at this point, if it wasn't clear at the start, I think personal budgets have tremendous potential.  My concern is that the chief aims of the personalisation agenda - empowerment and transparency - are undermined by practices like refusing to be transparent about how awards are arrived at, or only using the RAS if you think it will cut the cost of a care package.  I would also comment in passing that in discussions over the fair distribution of resources, we should not lose sight of the most significant problem facing social care today: the overall shortfall of resources to distribute.