By Ray Jones
Today’s National Audit Office report on child protection is timely but ultimately flawed, as it fails to look at the bigger picture behind the state of services.
Maybe this is an inevitable consequence of auditors digging down into the data detail without a helicopter view of the bigger landscape. Even so, it’s difficult to escape the conclusion that the report fails to see the wood for the trees.
The report tells about the tremendous increase in children’s social services and child protection workloads, pointing to a 94% increase in the number of children starting on child protection plans over the past decade. However, the NAO erroneously sees this as relating to the death of Peter Connelly in August 2007 when the major step change was in fact in November 2008 when the ‘Baby P’ media narrative started.
The data the NAO quotes shows how local authorities have increased expenditure on children’s social work in relation to child protection by 11% between 2012-13 and 2014-2015. This might be seen to indicate local councils trying but struggling to generate the capacity to deal with the considerable expansion in child protection activity.
It notes the growing vacancy rates in social work and the increasing reliance on temporary staff. Agency workers now account for 16% of children’s social workers, meanwhile 17% of posts are unfilled. These rates, because of cause and effect, are higher in services judged ‘inadequate’ by Ofsted. And the NAO report shows that more services have been rated by Ofsted as ‘inadequate’ or ‘requires improvement’.
But amongst all the data, an explanation for why this is all happening is not easily found in the NAO report.
The picture which might have – should have – been painted is one of more families getting into difficulty. Of more parents stressed and struggling with benefits and housing cuts and less help for children and families with the closure of children’s centres and the running down of youth services. Of the impact of the withdrawal of grants to community groups and voluntary organisations, such as Home Start.
This all contributes to more and more families being drawn into the child protection net and the difficulties social workers, councils and other agencies have in keeping up with year-on-year increases in activity at a time when local authorities face significant cuts to central government funding.
These efforts are then undermined by Ofsted raising the bar a council must meet to get a ‘good’ rating and having changed its terminology from ‘adequate’ to ‘requires improvement’. The result is that the NAO, comparing performance under two different inspection regimes, reports an overall deterioration in child protection services.
Possibly most alarming from the NAO report is their conclusion that there is no correlation between spend and quality of services. This might be seen – indeed it has been used in recent times by the government and the Department for Education – to argue that it is not the cuts at a time of increasing demand and workloads which is a major cause of the current pressures for many child protection services.
But there is an intervening variable, deprivation, which does correlate with spend and service pressures and performance as shown by separate recent studies by Bywaters, Hood and Bilson.
So what we have in the NAO report is largely an audit of data which, if read carefully, tells of a considerable increase in workloads with no similar increase in resources. But what we do not have is a clear view of the landscape behind the figures.
And unlike the recent House of Commons Education Select Committee report, which was critical of the government’s current disruptive set of children’s social work reforms, the NAO instead recommends the government focuses on improving data collection and actions that should be taken on that data (including intervening more quickly when data might show a council is struggling).
Scrutinising the figures is important. But the problem is that data without discussion and understanding is dangerous, and detail without perspective is limited in its prescriptive value.