Friday 6 September 2013

The real reason universal credit is in crisis

There is a titanic struggle about the way public services should be organised. It is only just beginning in a serious way, and it is below the radar of the mainstream media, but it is tremendously important.

It isn't the usual stuff - this isn't an argument between public and private.  It is between the conventional thinkers, committed to economies of scale and the manufacturing model, and the emerging critique – based on the ideas of the great quality guru W. Edwards Deming – which proposes something much more organic.

The news yesterday of the problems with the government's universal credit is some evidence for this.  In fact, it looks set to be the first great set-piece stand-off between the old and the new.

The leader of the insurgents, the system thinker John Seddon, confessed in his latest newsletter that he had temporarily abandoned the battlefield, blaming depression, and miserable that he was having so little impact on Whitehall.

But now he is back and as angry as ever, but he is quite right that Whitehall appears not to be listening to his absolutely fundamental critique of the way things are currently done.

From the very start, he has predicted disaster for the universal credit – not because the idea is wrong (it is tremendously important) - but because of the way it is being rolled out.

The fundamental mistake which the designers of the universal credit have made is the same one made by the NHS IT project – they have swallowed whole the mistaken assumptions of the IT consultants and are busily creating a system that can treat everyone online, as if they are on an assembly line.

What Seddon has shown is that this is the great flaw in the objective of digital by default.   It is a ruinously expensive fantasy because it is unable to deal with variety.

If a system can't deal with anything non-standard – and most of us are actually non-standard – then the inflexible system just bounces them around, creating costs and what Seddon calls 'failure demand'.  Failure demand can involve most of the demand on public services: it is people who are not being effectively dealt with, and keep coming back and back and back.

That is why digital by default always produces such unexpected costs. There is no reason why the universal credit should fail in itself, but this inflexibility condemns it to precisely the fate that it appears to be accelerating towards.

We are heading towards a new public service era which is characterised instead by a system which you might call human-by-default, because it is more effective and considerably cheaper.  The new era will be dominated, not by obsessive cost-cutting, but by revolutionary increases in effectiveness - which will, in turn, reduce costs.

But we are certainly not there yet.  It means recognising what human beings can do brilliantly and what IT can do brilliantly, and realising which is which. See more in my book The Human Element.

Unfortunately, the coalition appears to be stuck in the old assumptions, and Seddon's last newsletter described a depressing encounter with Health Secretary Jeremy Hunt, which nearly sent him back into the blues:

"I went to the minister for health’s presentation at a think tank a couple of months ago. I told him (Jeremy Hunt) that we’d been studying the NHS over the last couple of years and what we found was high levels of failure demand caused by fragmentation, specialisation, working to arbitrary measures and so on, and I suggested the way we should go would be to design a service that meets demand. He said he agreed with everything except my conclusion for, in his view, computer systems will fix the problem of integration. So there you have it. Policy amounts to what the minister will agree with. I think I’ll go back to my favourite mood."

Still some way to go, clearly, before the new era I described.  The trouble is that against Seddon and his handful of allies lies the immense marketing budgets of the IT consultancies.


3 comments:

Anonymous said...

"If a system can't deal with anything non-standard – and most of us are actually non-standard – then the inflexible system just bounces them around, creating costs and what Seddon calls 'failure demand'. Failure demand can involve most of the demand on public services: it is people who are not being effectively dealt with, and keep coming back and back and back."

If the vast majority of people do fit a fairly simple pattern (for whatever is under consideration) then a standard IT system may be appropriate - for those people only.

Non-standard situations might best be handled via manual systems instead of all-singing, all-dancing IT systems which all too often don't sing and dance but croak and fall over.

BrianW said...

It is an interesting argument: that 'digital by default' cannot deal with variety. Certainly, Universal Credit needs to deal with an amazing spectrum of cases. There are 6 different benefits, and each of these requirement dozens of pieces of information to assess one household (one claim covers the circumstances of teh whole household, don't forget). Fiendishly complicated.

Could Agile 2.0 have helped? I doubt it, and so does Ken:
http://bit.ly/Agile-2-point-zero-blog

Brian

Anonymous said...
This comment has been removed by a blog administrator.