WRITTEN ON December 18th, 2005 BY William Heath AND STORED IN Uncategorized
Kim Cameron blogs about the tax credits fraud
I wish I had some explanatory notes from designers who believe that hinging all government systems off national insurance numbers would be a good idea… I’ll be glad to post them along with this case study for their thinking.
Well, I saw the story, and Richard Allan’s post, but…come to think of it, I don’t really know what the lessons behind this incident are. Anyone know anyone who was involved?
And is there a WIBBI, apart from the usual make public services simpler & more navigable, and build IT systems professionally on a decent foundation of trust? We’re going to need to get more specific soon.
2 Responses to “Can we get inside the tax-credits fraud?”
Kim Cameron is ususally more than sensible. But on this one, he has missed the point in several different ways.
1. His headline talks about a two billion dollar fraud. The easy point is that he seems to imagine that the dollar is at parity with the pound. More importantly, he is confusing fraud and overpayment. There is a vast amount of overpayment in tax credits, as a result of its design and implementation. That’s not a good thing. But it’s not fraud. The article Cameron cites cites an NAO figure for fraud of £460 million – but even that is not identity fraud. By deinfition, it’s not easy to be precise about the numbers, but in other income related benefits, the best estimates are that identity fraud is less than 10% of the total.
2. All government systems are not hinged off national insurance numbers: very few are. That the system for deductng taxes due from earned income shares a reference number with the system for underpinning low levels of earned income seems neither suprising nor self-evidently inappropriate.
Less directly linked to Cameron’s comment, but there is an important process WIBBI here: let’s be clear what the problem is before rushing to solutions.
On the substance, the problem in this case is the result of the interaction of two weaknesses:
- the process by which DWP staff data was stolen (and the specificity of the group affected – one Jobcentre Plus region in one financial year suggests to me that the breach is more likely to have been of human rather than technical security)
- the process by which false claims to tax credit were accepted by HMRC.
The second of those is the more interesting, not least because in the short term there is a degree of trade off between making the process as unintrusive as possible for genuine applicatns while weeding out as many fraudulent ones as possible. Note, by the way, that zero fraud may not be the socially optimal outcome if achieveing it results in barriers which discourage genuine applicants.
My WIBBIs:
a). Admit that “targeted” (ie. means tested) benefits should be only for the tiny minority of exceptional cases. All other benefits should be automatic and “flat rate;” the normal tax system being used, where necessary, to recover money from the “better off.” This would greatly simplify administration and reduce unnecessary intrusion into people’s lives. It would also help restore the incentive to save.
b). Stop tinkering with taxes and benefits; allow procedures and IT systems to stabilize.
c). Tell the public the truth about this and other recent problems: Otherwise we’ll assume the worst. (Or blame the innocent!)
d). Tell DWP staff the truth and involve them in finding a solution: Staff still do not know the details and they find it very hard to trust DWP “management.”