HM Treasury unplugged - Government's IT late list
Why does it always go so wrong?
Analysis The Conservatives have helped expose, again, the systemic failure of Government IT projects with a seemingly trivial parliamentary question about costs and timescales at HM Treasury.
A written answer extracted by Theresa Villiers, shadow chief secretary to the Treasury, discovered that IT projects were running a total of 17 years late at HM Treasury under the leadership of Gordon Brown.
On the face of it, the answer provided an exhaustive list of 88 Treasury IT projects, about a third of which were running late. Altogether, taunted the Tories, the Treasury was running 17 years late with its computer systems. Or Brown, they might say, was 17 years too late.
No great surprise there, then. Yet a third of these projects are coming in under budget, not what you might expect of a tally of government IT projects, but perhaps in character for a department full of accountants. Most of the rest are either precisely or very nearly on budget.
One of the most notable thing about this tally is that it requires a parliamentary question for it to be published at all, and even then the bare minimum of information is conveyed.
This, and those projects in the list for which the Treasury has not been able to provide any information whatsoever, illustrate the lack of transparency and accountability in British public life.
Take, for example, the HM Revenue & Customs' modernisation of its PAYE tax system, called MPPC 2 (Modernising PAYE Processes for Customers). The Treasury answer claimed the project started in February, but had no idea when it was to finish, or how much it might cost.
MPPC 2 has been something of a mystery project that neither HMRC nor its prime contractor, Cap Gemini cared to discuss openly, especially when they were asked about rumours that it had been left hanging because of a lack of resources and snarl-ups with other HRMC systems, such as the notorious tax credits.
In the summer when we were asking about this, HMRC would say, we can't comment on work being done by our subcontractors, Cap Gemini, and CapG would say, we can't comment on work we're doing for our employer, HMRC.
Why so evasive? Surely, a bunch of accountants wouldn't let anyone start a major IT upgrade without any idea of cost or timescale?* That's like telling your builder to take your credit card and let himself out when he's finished the house extension.
Fortunately, another recent parliamentary question on government IT projects provides some context. On 4 September, in answer to a similar question by the Liberal Democrat MP Vince Cable, the Department of Health provided a tally as well. The only project for which the department had no clue of when it started, when it would end and what it might cost was the infamous National Programme for IT, the IT industry's answer to the Millennium Dome.
The DoH answer waffled that NPfIT didn't really having a start or end date because it was sort of, well, "substantial", being planned on the fly, "incremental", and "providing increasingly richer functionality over time." **
Similarly, HMRC told The Register in June that MPPC 2 didn't have a deadline because it didn't need one. It was a special project. It had a new sort of deadline, which was sort of amorphous. Its size and complexity had something to do with this, but it was a little vague.
This is only part of the story, however. NPfIT faltered because it was imposed from above, without reference to the clinicians who were to use it. Connecting for Health, the organisation responsible for NPfIT, admitted that if it had consulted the intended users of the system more widely and included their views in its design, they might have a better idea of what it was doing.
It was trying to be too big, too clever, and had tried to impose its world view on too many people. It's a common problem with computer systems, as was the case with the Child Support Agency computer system, which made a major contribution to the death of its own agency, and HMRC's tax credits system, which had imposed a rigid and impractical method of working on staff.
Sponsored: Hyper-scale data management