Think tank to UK government: You can't build the future on systems from the past

Flagship tech projects such as the ID card scheme are at risk of failure unless the UK government changes its approach to legacy systems - which evidence shows is getting worse, a new think tank report claims. Re:State, a non-partisan policy unit focused on public service reform, says much of the government's ambitions for digital services and efficiency depend on "modern, interoperable systems." However, the problem of legacy systems is underestimated, it claims. "In Westminster the money doesn't get prioritized for tech, and so behind the scenes successive governments have neglected to fix many dangerously outdated systems, leaving a ticking time bomb for future generations to defuse," said Joe Hill, co-author of the report, director of Strategy at Re:State and former Treasury civil servant. Examples are not hard to find. They include problems migrating the Police National Database to the cloud, the scandalous data breach revealing the names of Afghan informants, and a creaking farm payments system. The problem lies in departmental control of legacy system remediation and the funding model for those projects. The Re:State report, From legacy to leadership [PDF], says that funding comes in two forms: crisis funding or maintenance funding. "Systems aren't transformed unless they fail in substantial ways. The result is that the gap between what systems can do and what services require widens each year. Departments fall behind with out-of-date technology stacks by relying on aging platforms that constrain service design, data use, and automation, which leaves them with ever more catch-up to play at a later date as operational urgency rises," it states. Much of the report relies on data from the State of Digital Government Review 2025, which found lost productivity from legacy IT cost 4-7 percent of annual public sector spending, holding back both productivity and public satisfaction. That review found the proportion of legacy systems in central government was around 28 percent. It ranged from 10 to 60 percent, depending on department, and had increased by 26 percent since 2023. Of those legacy systems, 22 percent were considered "red-rated," meaning they carried risks judged both highly likely and high impact. The proportion of red-rated systems had also increased. The scale of the problem and its embedded nature means that continuing with a department-led approach to tackling the legacy system problem won't work, the paper argues. Because there is little reward for prioritizing reduction in reliance on legacy systems, departmental leaders tend to focus on broader transformations, which come with more incentives and rewards. Budgeting is also a problem. Tech funding is awarded based on projects, rather than services, which makes underinvestment likely in two ways. "Firstly, because core operating costs of existing legacy technology have to constantly be reapproved as projects, making it easier to negotiate technology investment down in favour of other areas. And secondly, because it allows policymakers to plan additional investments in new technology like AI without thinking about investing in the underpinning services, which often have legacy IT components," the report adds. A new central government "Digital Modernization Taskforce" with a mandate to reduce systemic legacy risk and embed prevention, is one solution proposed. The report also proposes to tackle funding. "When central government investment is available for a particular kind of spending, such as legacy IT, interviewees for this paper felt that could disincentivise departments to make their own investments instead of 'waiting to see if [the Department for Science, Innovation and Technology] will fund the risk instead,'" the report states. "Instead, the Taskforce should adopt a 'match funding' model - using centrally allocated funding at the next Spending Review to match the amount that departments put into their own legacy IT transformation projects, in order to speed those up." The report has five other ideas for how the government can escape the deepening quagmire of legacy IT, including new approaches to procurement and supplier management. Welcome they might be, but with the government seemingly fixated on headline-grabbing announcements, only an optimist would expect to see them in action. (R)