Paul Cook is an independent consultant who has run
social services finance at five major authorities. He is the author
of the Chartered Institute of Public Finance and Accountancy’s
finance guides on housing, housing benefit, supporting people and
partnerships. He has also been director of finance at Westminster
Council and chief executive of Daventry Council.
Dealing with unit costs is an integral part of managing
performance. Appliances issued per thousand population, runs per
wicket in bowlers’ cricket averages, percentage of trains leaving
Euston spotted per trainspotter and so on.
Social services unit costs are serious business. The government
likes to report that national performance onÊkey social care
indicators is improving yearly.ÊThese national average
indicators are built up from the performance assessment framework
that each social services department must complete locally. The
framework is used to assess delivery of key targets in the NHS
Councils’ star ratings include an element of how their unit costs
look. Higher star ratings contribute to improvements in the
council’s overall comprehensive performance assessment.
A higher social services star or overall rating also creates more
freedoms and flexibilities at local level. Particularly useful
freedoms are the ability to use ring-fenced grants for any purpose,
and fewer inspections or even an inspection holiday
Textbook performance management expects managers to set unit cost
targets and see how actual results pan out. Unit costs should be
compared with those of similar organisations. The upper quartile
should be aspired to. Or, in the case of the performance assessment
framework, often the lower quartile or even the safe middle ground
between the two.
Unusually high or low unit costs should prompt more detailed
investigation and analysis. Enlightenment and progressive service
changes should ensue as the unit cost cuts away management fuddle
and reliably measures the delivery of those infamous “efficiency
Most councils practice limited textbook performance management. But
there is no doubt that getting unit costs right for performance
assessment is essential to improving or maintaining that all
important star rating.
Achieving robust unit costs.
Every financial year the Chartered Institute of Public Finance
(Cipfa) produces an edition of its Best Value Accounting Code of
Practice – England and Wales. This includes a standard expenditure
analysis for social services. This has been developed over several
years to achieve consistency between councils’ accounts and
statistical returns. If you do not apply this analysis, your costs
will not be comparable. The overriding principle of the standard
expenditure analysis is to break down expenditure over client
groups (children, older people and so on). There are three points
First, Cipfa requires social services to bear its fair share of
central overheads. Support service and management costs within
social services need to be allocated out to individual
Second, there are strict rules about the bases of allocation. These
should be applied consistently in budgets and accounts. Don’t let
the allocations be fudged to produce plausible unit costs.
Third, when completing Department of Health returns, read all the
instructions thoroughly. Often there are changes from year to year.
These usually affect which clients and which costs should or should
not be included.
Here are some common errors in coding expenditure or
producing activity data, identified from Cipfa’s 2003 accounting
code of practice. Getting these wrong can overload or underload
your costs and finance data:
- Service strategy and regulation is a limited category. It must
be restricted to the director’s office and the complaints
- Sheltered employment does not count as social services for
- Supporting People is a housing service. But you can include
some Supporting People expenditure in social services if it is
incidental to the main service in question. But you are entitled to
strip it out of personal social services unit costs.
- Asylum is a service in its own right and should not be included
in the cost of other client groups.
- Clients become older people for statistical purposes on
Combining finance and activity data is where a lot of councils
start panicking about their unit costs. A unit cost is the result
of a finance numerator and an activity divisor.
For example, annual expenditure on older people’s residential
placements divided by number of placements weeks equals cost per
week of a residential placement.
Anomalies in either element of this calculation – finance or
activity – will lead to anomalies in the result. Often it is only
when the two are put together that it is clear one of the figures
is bad. For example, “I drove for 100 miles” seems a reasonable
statement, as does “I drove for 20 minutes”. But “I drove at
300mph” does not.
Here are a few tips to avoid unit cost nonsense:
- Put together both elements of the unit cost calculation well
ahead of the return date. This will allow time to resolve
discrepancies and check out the figures. It is surprising how many
councils leave the joining of activity and finance data until a few
days before the return is due. Panic then ensues when the figures
- In many social services directorates, finance and activity
input are collected by different sections. This puts a premium on
good communication. Both teams need to own the unit cost
- Be self-critical and open about your own data quality. Often,
when people are quizzed about their bizarre results, it becomes
apparent they know only too well what is wrong.
- Similarly, recriminations, snide observations or open onslaught
on somebody else’s data quality really don’t help. At worst, this
will lead to teams attempting to hide their own misgivings on data
There is a series of short cuts that can be taken to test for
- Check the correct budget figures and the correct final accounts
figure are being used.
- Ask for a map showing how the total cost in the approved budget
or final accounts is mapped into the boxes of returns. Inability to
produce this is a sure sign things are wrong.
- Ask for a map showing how the activity data entered in boxes of
returns is derived from approved systems and databases.
- A set of costs in a return may look plausible. But how do the
figures look taking a succession of years together? If the unit
cost of a placement oscillates wildly from year to year with no
service reason, it is a fair bet that some or all of the unit costs
are wrong. The same point applies to service activity data.
The twice yearly delivering improvement statement demands a raft of
finance and unit cost information. It is submitted electronically.
The workbook you are required to complete prompts you if your entry
shows an abnormal variation from previous returns. The spring 2004
delivering improvement statement was due on 31 May. The autumn
version will probably be required in November. A more detailed
expenditure analysis based on final accounts is the form PSS EX 1.
This should be with the DoH by 31 July.
Unit costs are important because they have a big influence on
directorates’ star ratings. This article examines what they are and
why they are important. It also considers the use of unit costs by
the Commission for Social Care Inspection and the performance
assessment framework. It also looks at the dos and don’ts to
achieve robust unit costs, as well as short cuts to test a set of
unit costs for accuracy.
The rest of the series.
- 24 June: Grants are a big portion of social services income
but they are complex beasts to administer. A look at how to keep
track of your grant income.
- 1 July: Weak forecasting and monitoring are the main financial
danger areas for social services managers. How can you avoid
Contact Paul Cook at firstname.lastname@example.org