Just another few billion down the drain in government IT

Computer says “not on your nelly, mate”. quinn.anya

A report from the Public Accounts Committee has revealed that the National Programme for IT in the NHS has cost the taxpayer close to £10 billion, despite having been abandoned.

The committee’s chairman, Conservative MP Richard Bacon, has described the project as “one of the worst and most expensive contracting fiascos in the history of the public sector”. However, it will almost certainly not be the last fiasco of its kind, unless serious changes are made in the way the government handles its IT ambitions.

The NHS digitisation saga dates back to 2002, when an IT project aimed at revolutionising the NHS was born. In 2003, a contract was signed with outsourcing company CSC to develop Lorenzo, a system for patient care records. Although the overall National Programme for IT was eventually axed, parts of it continued as separate projects. But a decade on, the Public Accounts Committee says not a single NHS trust has a fully functioning Lorenzo system. It concludes that the public continues to pay the price for the mistakes made by the Department of Health and its contractors along the way.

Just a few weeks ago, another report on a failing government IT-enabled project made for equally depressing reading. In this case, the National Audit Office documented the remarkably unmitigated lack of progress made by the Department of Work and Pensions as it implements a Universal Credit system, simplifying six benefits into one.

The government planned to spend £2.4 billion from 2011 to 2023 to implement Universal Credit. Most spending so far - £303 million - has gone towards contracting for designing and developing IT systems. Yet there still appears to be no agreed design or IT architecture.

Referring to this latest saga, the Public Accounts Committee has warned that the government is still a long way off learning lessons from its past mistakes in IT contracting projects such as Lorenzo.

Far from having reflected on the disasters of the past, the DWP ploughs on with a project that so far looks unable to support a national roll-out without much further work and investment. The current IT system is unable to deal with fraudulent claims and there has been a £34 million write-off of IT systems already, with more anticipated.

These cases reflect a longstanding problem with IT projects across government. Recall the chequered history of IT and outsourcing at the Department of Social Security, the Ministry of Defence, the Home Office, and HM Customs and Excise, and you get the sense that something is rotten in the state of Denmark.

What needs to be done, then? On this, the NAO provides a good report, sound advice, but is unlikely to make a material difference. Focusing on one project, it can only hint at five deeper more intractable issues at the heart of government.

First, politicians regularly treat IT not only as an administrative tool, but also, as a “fire-and-forget missile”. IT in government can be transformational; in practice it is perceived by politicians as a relatively easy-to-implement cost saving device that reduces headcount. However, they often cost huge amounts of money, spread over many years. There are perverse incentives at work here: all the benefits for ministers are upfront, with the price paid after they’ve gone.

Then there has been the dismantling of the public service ethos through an often ideological, all too frequently misplaced faith amongst the political class, encouraged by advisers, consultants and suppliers, in the superiority of “marketisation”, and “businessisation”. A related issue is the strangely strong belief in the ease with which private sector practices can be applied to distinctive and very different public sector contexts and issues.

A further factor has been the long-term erosion of project management and IT retained capabilities within departments. For years, there has been a desperate need to bring in more public service managers who are able to understand the technical and process architecture of their departmental domains, and are able to manage external suppliers. The Department of Work and Pensions, for example, agreed to use an unfamiliar agile methodology in developing the benefits system, while the implementation team grew to a less than agile 1,000. Planned savings are already down by £500 million. All post-2015 milestones are “under review”. You get the idea.

Then there is the remarkable faith in large-scale, long-term outsourcing with big suppliers. Companies like Fujitsu are not the only option for IT projects and are in fact probably not the best strategic choice for under-resourced government departments. Recently, there has been talk of trying to get smaller businesses involved in these projects but little progress has been made so far.

Finally, there is the blame game. In the larger context, all involved parties have become highly adept at setting up major projects so that blame can always be allocated elsewhere, but no one is ultimately responsible. This tacit conspiracy has been exacerbated by the rapid growth in outsourcing, use of consultants, deregulation, and privatisation. It makes for plenty of hiding places and indicates a serious lack of leadership at the heart of government IT. Until these factors change, we can certainly expect more horror stories like Lorenzo.