Fraud, error and administrative cock up are a real problem for our benefits system. Despite this we sink to the level of ‘debate by headline’ where the prejudices and preferences of different sides to the argument are paraded like a political version of a 1960s Brighton Bank Holiday.
What follows is an attempt to put a little perspective into the argument – to escape from silly ad hominem and straw man arguments and actually talk about the issue – fraud, error and administrative cock up in our benefit system. And let’s be clear this stuff is there – benefit fraud isn’t the invention of frothing right-wing pundits writing in the Daily Mail. Every Government in my memory has, at some point, announced a ‘crack down’ on benefit fraud (albeit mostly to get a headline or two).
So what’s the deal? How big an issue is fraud and what should we do about it? Well much has been said about it – anecdotally and statistically and the new Government is no exception:
“We are looking urgently at different options for reform. Tougher penalties for
fraud, more prosecutions, encouraging those who know fraud is taking place to come forward, and making greater efforts to reclaim money that's wrongly paid," the prime minister wrote. "We will look at all these things and more. Including, for example, using more information from third parties such as credit referencing agencies to identify circumstances which are incompatible with the benefit claim. I have asked Iain Duncan Smith [the work and pensions secretary] to draw up an uncompromising strategy for tackling fraud and error which we will publish in the autumn."
Cue frothing media debate and a stream of absolute nonsense from pundits and politicians on all sides – from scaremongering about the use of credit referencing data to hyperbole about ‘bounty hunters’. Plus, of course, the compulsory “it doesn’t really matter, it’s just a drop in the ocean of Government spending” and “why are we worrying about benefit fraud, what about tax dodging?”
So, from me, an attempt at a more balanced analysis!
Firstly, benefit fraud is a problem – somewhere between £1.1bn and £5.5bn worth of problem. And it is compounded by the additional problems of overpayment, underpayment and ‘claims in error’. Once we shake all this stuff down there’s something around £10bn of spending involved – which is a lot of people’s taxes however you want to slice them (I would net this off against under-claiming of benefits except the Government already does that in its projections on public spending so to do so would be double counting – just as is the case with estimates of tax revenue).
The problem is that ‘clamping down on fraud’ costs money – you need to employ armies of fraud investigators and I’m sure they don’t come cheap. So there must come a point where further clamping down on fraud – however morally justified – actually costs the taxpayer more money than allowing the fraud to continue. Unless, of course, you can find another way to ‘fight fraud’.
Well there are two important considerations here – what negative incentives can we build in to prevent fraud? The obvious one is to catch more fraudsters (this influences both the fraudster caught and the whole market since the more caught the less incentive to cheat). Except that we’ve already spotted how this doesn’t make fiscal sense. Unless we shift the Government’s risk (in employing fifty grand’s worth of investigator) into the private sector. This is called by some ‘bounty hunting’ but it might be more sensibly named speculative private investigation – you bring evidence that leads to a successful prosecution for benefits fraud we pay you an agreed amount. This amount would represent the median cost of a benefit fraudster to the public purse – the only risk the government faces is that the investigator behaves badly. And, as far as I know, we have other laws to deal with that!
The other approach is to make is less profitable for the fraudster to take the risk – either because of the increased chance of being caught or else because the balance of risk and reward isn’t good enough. In a less complex system based on less information, fraud would be less profitable (and error less likely – addressing a related problem). At present we have a complex system that is relatively easy to game which results in a great deal of abuse – some of which is openly fraudulent while the rest is what might be called ‘pushing the envelope’.
As it stands – and given the number of claimants and the amounts of money involved – there is no chance of any Government making much of an impact on fraud. And with what we might call “fiscal ullage” running at a maximum of 2% and probably less, it makes little sense (beyond headline moralising) to run such programmes and campaigns. Unless of course you take my advice and shift collection risk into the private sector – which means fraudsters will either get caught or else the problem will disperse because the risks are too high. Simple really!