1.  The reported amount of benefit fraud has reduced, but the reported amount of customer and official error has nearly doubled in the last five years to almost £2 billion. The current extent of error is unacceptable and has contributed to the continued qualification of the Department's accounts for the past 19 years. While estimated fraud has reduced to £800 million—around 0.6% of total benefit expenditure—official error rates for benefits such as Jobseeker's Allowance are still too high at 2.1%. To prevent official error increasing, the Department needs to target its training and compliance checks on those local offices and benefits which consistently have the highest error rates. Despite efficiency drives, the Department must not allow staffing cuts to have a detrimental effect on the fight against fraud.

2.  Fraudsters increase their gains if they target benefits that act as passports to other forms of financial support. Many benefits act as a gateway to further benefit entitlement and failure to prevent invalid claims in these increases the risk of fraud in other benefits. Since our predecessors' 2003 Report, the Department has strengthened the checks carried out at the outset of a benefit claim. It should pay particular attention to the internal controls around those benefits where there is a risk that initial entitlement to one benefit could be used to obtain others to which the customer is not eligible.

3.  The Department is working more closely with local authorities to safeguard Housing Benefit from fraud and error. Jobcentre Plus is piloting a scheme under which its offices act as customer service centres for Housing Benefit, currently administered through 408 local authorities. When evaluating the pilot scheme, the Department must take account of the impact the new arrangements will have on the capacity of Jobcentre Plus offices to deliver its existing services. It must mitigate the risk of overloading existing staff with additional and complex benefit rules that may lead to more official error in other benefits.

4.  Organised crime continues to present a serious threat to the benefit system and needs careful management. Organised crime is often complex and well planned. In 2006-07, the Department took some 900 cases of organised fraud to court, and around 70% of those cases which were prosecuted ended in an actual or suspended custodial sentence. Where the Department detects organised attacks on the system, it must take a firm and co-ordinated approach against these fraudsters, using both internal resources and working with external partners such as the police.

5.  In 2006-07 only £22 million of £339 million known fraud debt was recovered. Counter-fraud measures focus on overpayments identified, rather than monies recovered. The Department should set recovery targets for overpayments due to fraud and underpin these with a recovery strategy. The overall deterrent effect of the Department's counter-fraud work will ultimately be reduced unless money is recovered.

6.  The Department has taken too long to improve its management information systems. Despite being told in 2003 that the Department was developing a better management information system, called FRAIMS, it was only in February 2008 that the new £65 million system was rolled out. Moreover, it is not sufficiently comprehensive. The Department needs to assess the effectiveness of this management information system as it is rolled out, and evaluate the future impact on those counter-fraud activities which currently are not served by the system.

7.  The cost-effectiveness of counter-fraud activity needs to be more routinely measured. The Department reported that it spent some £154 million combating fraud in 2006-07. It needs to understand the relative cost effectiveness of its different counter fraud actions in order to determine where best to target its resources. The Department should develop a robust methodology for measuring the cost effectiveness of its work, drawing on its new management information system and the approach used in the Comptroller and Auditor General's Report.

8.  Some 200,000 cases of potential fraud were investigated in 2006-07, where the Department considered there to be a high probability of a prosecution, but only 7,500 were taken to court. Potential fraudsters will not be deterred if the Department is not seen to be taking firm action where there is good evidence that fraud has taken place. The Department's 90% success rate at court suggests that they are selecting cases to prosecute which are likely to have a low risk of failure, but its lack of adequate management information makes it difficult to determine what other cases the Department should be prosecuting. The Department should analyse the outcome of prosecutions by case type so that it can make better informed decisions about the cases it should prosecute.

9.  The Customer Compliance approach is potentially a good way of handling lower level frauds but there is a risk that it may erode the message that the Department is tough on fraudsters. The Department should review the cost effectiveness of Customer Compliance now that it has entered its second year of operation. This should consider the costs, performance and effectiveness of the approach, and also examine whether there are any signs that the model is undermining the Department's hard message on prosecution of fraudsters.

10.  The Department needs to take proper ownership of these recommendations and put in place arrangements to implement them. Once the Department has published its Treasury Minute, setting out its intended actions in response to the Committee's recommendations, it should give responsibility for overseeing their implementation to a single senior official.