The main argument against innovating to improve the workings of government is that the legacy systems work - so don’t touch.
But are the big systems - the ones that handle tax and benefits - working so smoothly that inventiveness and experimentation around them would be too risky?
Today’s report of the Public Accounts Committee says the extent of fraud and error has resulted in the accounts of the Department for Work and Pensions being qualified for the past 22 years.
“The extent of fraud and error has resulted in the Department's accounts being qualified for the past 22 years.
“Sir Leigh Lewis, the Permanent Secretary for the Department until his retirement in December 2010, told us that the failure to lift the qualification was the 'biggest single disappointment' of his five years leading the Department.
“The new Permanent Secretary, Robert Devereux, told us he was now committed to removing the qualification on the accounts.”
It’s a similar story at HM Revenue and Customs. The National Audit Office said in its report on the 2009/10 accounts of HMRC:
"The levels of error and fraud are material within the context of the £27.3 billion spent on tax credits. As this expenditure has not been applied to the purposes intended by Parliament and does not conform with the requirements of the Tax Credits Act 2002, the Comptroller & Auditor General has qualified his opinion on the regularity of the tax credits expenditure reported in the 2009-10 Trust Statement."
The former head of a DWP department said recently that he would have lost his job if citizens had gone unpaid. But that’s highly unlikely even if systems and processes were disrupted. It’s more likely that people would be paid - but perhaps the wrong amounts.
Which is what happening at the moment to a considerable extent.
If things cannot get much worse, what could be so bad about experimentation and innovation? Everything to gain, surely?