Ed Balls, the children’s secretary, has said that parents who look after each other’s children will not have to undergo criminal record checks and take childcare courses to make their arrangements legal.

He told MPs yesterday that he had agreed with Ofsted that this kind of arrangement was beyond the scope of their childcare inspections. His decision was prompted by an Ofsted inspector telling two police officers that they were breaking the law by taking turns to look after each other’s daughters while they worked shifts at Aylesbury police station.
 
While this is an overdue outbreak of common sense, it raises some awkward issues for the Independent Safeguarding Authority, which plans to vet 11.3 million adults, including many who take responsibility for children as voluntary helpers “frequently or intensively”. Since frequently means once a month or more over a period of three months, such reciprocal arrangements would clearly fall within it.
 
Last month Mr Balls asked the authority to look again at these definitions. But his advice over the Ofsted inspections seems to prejudge this review, and suggests that the authority is going to have to trim its massive list of potential criminals. (On September 18 it promised to tell me within 10 days how it reached the total of 11.3 million, or refer me to a document explaining it. I have yet to get a reply ... 16 working days later. But I dare say they’re busy.)
 
The real problem, however, is not Mr Balls’ embarrassment, but the implications of what he has just done. The majority of offences against children are carried out by parents or family friends, not by total strangers. If friends and acquaintances are excluded from the safeguarding system (which is the implication of his instruction to Ofsted), it will be rendered even more ineffective that it already seemed likely to be.
 
The Home Office published a paper in 1998 by Don Grubin (Police Research series paper 99) which summarises the evidence on sex offences against children. He  said: 

“Of those who reported some form of sexually abusive experience as children, 10 per cent said that the perpetrators were relatives (parents, siblings, uncles/aunts, cousins or grandparents), 46 per cent someone known to them (either an adult or peer), and 44 per cent  a stranger. However, if incidents of “flashing” are excluded, then 13 per cent of the perpetrators were relatives, 68 per cent known and just 18 per cent strangers.” 

Much the same is true of homicide. The majority of children who are killed die at the hands of family or friends. Less than a quarter are killed by strangers, according to the NSPCC.
 
So a vetting system that excludes family, unpaid volunteers, and informal arrangements like that of the two Aylesbury policewomen will exclude something like 75-80 per cent of the likely perpetrators of murder or serious sexual assault against children. That, I guess, is precisely why the Independent Safeguarding Authority included such people. (We should be grateful they drew the line at vetting families, even though they are the most likely perpetrators.)
 
The truth is that anybody familiar with the pattern of offending against children would have realised that a safeguarding mission large enough to vet the majority of likely offenders would be too large and unconscionable to contemplate, and one that was small enough to gain public acceptability would miss the great majority of possible offenders, and would be no better than existing arrangements. In other words, the game wasn’t worth the candle.
 
But I don't recall hearing any of these arguments from childcare charities, MPs or peers as the bill went through Parliament. It will be interesting to see if Mr Balls can manage to wriggle out of this one.