Recommendation 3

Previous efficiency programmes have over-promised and under-delivered. ‘Optimism bias’ is a long-standing problem within government where project plans are impractical and unrealistic, and do not always achieve the expected objectives. We have seen this across a range of programmes, such as Shared Service Centres and, more recently, the Defence Equipment Plan 2020–2030. These problems occur due to poor data on costs and performance, but also because departments and central government do not consistently challenge existing plans. The Treasury and Cabinet Office told us that “test-and-learn pilots” gather programme data to give departments 6 Efficiency in government the flexibility to react when programmes do not meet their assumptions. However, government needs to be more agile in responding to these results: currently too few schemes are halting early enough after showing signs of poor progress. These issues could re-emerge if government prematurely over commits to efficiencies based on changes that have occurred during the pandemic which may not endure, such as reducing the government estate as a result of staff working from home. Recommendation: HM Treasury need to ensure plans are subject to adequate challenge, testing the realism of departments’ expected savings, and considering the use of pilots where appropriate.