Summary
- MPs say the government cannot yet substantiate £4.3 billion of projected shared-services benefits.
- Five departmental clusters are consolidating HR, finance, payroll, and procurement systems.
- Standard software will not produce common services without reliable data, ownership, and departmental participation.
The government’s attempt to consolidate HR, payroll, finance, and procurement systems across Whitehall risks becoming another expensive transformation programme without an agreed timetable, convincing benefits case, or clear ownership, according to the Public Accounts Committee.
The Public Accounts Committee says the Cabinet Office has not demonstrated how its shared-services strategy will generate the £4.3 billion of benefits it claims, particularly because the figure does not fully account for programme costs.
At least £1.15 billion has already been committed, while the planned start of a major delivery phase has been delayed by five months without a satisfactory explanation. MPs identified complicated governance, inconsistent departmental participation, delayed data preparation, weak oversight of dependencies, and no clear owner for the overall result.
The committee’s report also questions the evidence supporting the largest of five departmental clusters, where Capita has been contracted to provide services for organisations including the Department for Work and Pensions, Home Office, Ministry of Justice, and Department for Environment, Food and Rural Affairs.
Common software cannot create common government
The principle behind shared services is straightforward, because departments carrying out similar administrative work should not all maintain separate systems, suppliers, data models, and support organisations. Common platforms can reduce duplication, improve reporting, and allow civil servants to move between organisations without encountering entirely different processes.
Whitehall is nevertheless attempting to standardise institutions with different histories, employment arrangements, financial processes, statutory duties, and legacy technology. Where departments retain distinct processes beneath nominally common software, the result can become a collection of expensive customisations rather than a shared service.
Data preparation is central to the programme because payroll, procurement, finance, and HR systems require accurate records, consistent identifiers, agreed classifications, and defensible access controls. Migrating poor or incompatible data into a replacement platform does not resolve the underlying problem and may make mistakes harder to trace across a larger estate.
The committee found that the Cabinet Office’s dashboard cannot compel departments to provide reliable information, which weakens its value as a management tool. The organisation coordinating delivery therefore lacks direct control over the departments whose participation determines whether the strategy succeeds.
Savings require a defensible comparison
A claim of £4.3 billion in benefits requires a credible comparison between current costs and the future operating model. The baseline should include licences, implementation, migration, contractors, internal staff, data remediation, support, change management, decommissioning, and the cost of running old and new systems together.
Projected benefits can otherwise become an accumulation of gross savings that overlooks the expenditure required to obtain them. A department may reduce one software contract while paying for integration, consultancy, additional controls, or temporary operating capacity elsewhere.
The Capita contract creates another assurance issue because the supplier is already responsible for several prominent public-service arrangements. The committee has asked whether the price was substantially below the modelled cost and what safeguards demonstrate that the promised service can be delivered.
Outsourcing does not transfer responsibility for service design or data quality. A provider can operate agreed processes, but departments still need to define outcomes, protect sensitive information, manage exceptions, and retain enough expertise to challenge performance or move elsewhere.
Workforce consequences extend beyond an efficiency figure. Shared services can remove duplication, centralise specialist work, and change how civil servants obtain support, whereas poor implementation can generate extra manual work, unresolved tickets, payroll errors, and local processes created to compensate for weaknesses in the common platform.
The committee has challenged the current delivery structure rather than the principle of shared services. The Cabinet Office can rebuild the programme around credible costs, ownership, prepared data, and firm departmental commitments, or continue with a strategy whose benefits remain much more precise than its operating arrangements.
Whitehall has no shortage of enterprise programmes founded on a reasonable idea. The recurrent failure comes when common software is treated as a substitute for common decisions, leaving the replacement platform to inherit the fragmentation it was purchased to remove.










