GP Mary Hawking has commented on an important point made in the Channel 4 “Dispatches” programme about the murky “clearance” process by which the public spending watchdog, the National Audit Office, agrees its reports with the department or agency whose work it has scrutinized.
Ostensibly the clearance process is not murky at all. It allows MPs who sit on the House of Commons’ Public Accounts Committee to question civil servants on reports of the National Audit Office [NAO] without arguments over the facts.
But the clearance process also allows a minister or the head of a department to stop an NAO report being published: they simply refuse to sign it.
The NAO’s report on the NHS’s National Programme for IT [NPfIT] was delayed so long that some auditors were considering setting a precedent by publishing it without sign off by the Department of Health.
In the end the final NAO report on the NPfIT was published without the criticisms that had been in the draft versions. The final report was so delicately complimentary about the Department of Health and NHS Connecting for Health, which runs part of the NPfIT, that MP Greg Clarke, then a member of the House of Commons Public Accounts Committee, commented on its tone at a hearing on 26 June 2006.
“In a year on the committee I have read 62 NAO reports. This is easily the most gushing,” he said.
In 1998 the NAO finally published a report on “Read Codes” that the Department of Health had delayed by about a year.
Now, a reader of this blog, Mary Hawking, asks:
“Is there any way to find the anticipated publication date of NAO reports, and is there any indication in the difference between scheduled and actual release dates of the degree of departmental amendment concerned before the report could be signed off?
“Can we (the voters) safely assume that if an NAO report is delayed, it contains criticisms which Sir Humphrey (and the Minister) would find embarrassing?”
I apologise for making some remarks that overlap in parts with a separate blog entry “Channel 4 Dispatches – Is the National Audit Office a mere critical friend to government?”
There is no easy way to find out whether an NAO report has been delayed because it doesn’t announce publication dates. The NAO indicated to Computer Weekly in August 2004 that it would publish a report on the NPfIT in the summer of 2005. But it did not commit to any date. In the summer of 2005, it indicated the NPfIT report would be published in the Autumn of 2005. In the end it didn’t appear until June 2006.
The NAO is now working on a second report on the NPfIT. But it may not be published for more than a year.
My view is that the NAO should check a draft report with the department or agency but should not negotiate for a signature. When an NAO report is complete, it should be published as soon as possible, ideally without a signature.
The NAO would oppose such an approach, as would departments and agencies. They prefer the present position in which the NAO has a qualified independence. By agreeing its reports the NAO and the department or agency present a united front to the Public Accounts Committee. So at hearings of the committee there is rarely a clash between the NAO and civil servants over the content of its reports. But such arguments would be a useful way to extract the truth, far better than a pretence that conflicts don’t exist.
As things stand, with consensus between auditors and the department being audited, there is a danger the NAO will report on transient aberrations, not fractured foundations. It’s easy to reach a consensus on things that don’t really matter, that won’t result in any radical action.
This is why in some cases the clearance process doesn’t always work in the interests of the departments or taxpayers.
The way the process operates now is that the NAO produces a draft report, a sort of MoT test on a project or projects, a policy undertaking or a financial investment by the department. It may find defects that are serious, not easily fixed and dangerous.
The Department’s PR-minded specialists are likely to counter the NAO’s draft findings, probably by producing a third-party report which finds there are no serious defects. Staff at the NAO staff may have doubts about the report’s independence – it could have been written by a consultancy that does a lot of work for the department.
Caveats, assumptions and extrapolations abound in the consultancy’s report. It’s largely speculation, and it’s based on facts provided by the department. The NAO has to accept the report or dispute its provenance. The NAO abhors confrontation with other public servants. So its final report questions the scuff marks on the seats, the dents in the bodywork and the broken light-bulbs. Its report mentions little or nothing that has profound ramifications.
Reports by the NAO on HM Revenue and Customs, for example, have produced evidence of dents and scratches, but have been silent on the department’s fitness for purpose.
There is also a penalty in time on the NAO’s reaching agreement with the department being audited. It means that an NAO report will often be out of date by the time it is published. This enables the heads of departments and agencies to tell the Public Accounts Committee that things have changed since the NAO’s report was published. They will say that the NAO’s recommendations have already been carried out. Whether this is true or not will not be known to the NAO’s representatives at the hearing of the public accounts committee. Neither will MPs know if the department’s head is telling the whole truth. Accountability and transparency will be diminished.
So the clearance process is murky. It doesn’t serve the interests of departments which would benefit from robust independence and regular external scrutiny. Sometimes it takes an external auditor to say things staff have avoided telling their senior managers, or their senior managers have avoided telling ministers.
The NAO would argue that it is completely independent of government, that it agrees only the facts with departments and agencies, not its conclusions or recommendations. But a comparison of the draft NAO reports on the NPfIT and the final version shows that more than facts get changed. Critical comment by the NAO was removed. Marketing material was added.
A collegiate, consensual relationship between department and the NAO may let through some light. But reading an NAO report can be – at times but not always – like going to an art gallery and looking at a picture through a telescope.
Unless there is comprehensive disclosure of what has gone wrong and why, lessons may not be learned. If a plane crashes and the real cause or causes do not emerge other aircraft may be doomed by a lack of public disclosure of the truth.
Dispatches – Nice Work If You Can Get It – broadcast 24 September 2007