Skip to main contentSkip to navigationSkip to navigation

Don't just do something, stand there

This article is more than 17 years old
Daniel Davies
Constant reform has left the public sector in a shambles. It's time to stand up for the status quo.

With Accenture walking away from a number of NHS contracts yesterday, iSoft in pretty serious disgrace and the actual NHS IT system upgrade apparently stranded on a rock, it has been a bad few weeks for the myth of private sector know-how in big public sector projects. Perhaps we might do a bit better by looking back to the last big success story of public sector management of a big technology installation.

You have to look back quite a long way, to 1996, when Crest went live as the electronic processing system of the London Stock Exchange. Crest is now part of Euroclear and a private company, but at the time it was a subsidiary of the Bank of England, who had stepped in when its predecessor, Taurus, had collapsed in 1993.

The collapse of Taurus was a huge embarrassment for the London market; there was a public inquiry into its failure which ended up concluding that it had been fundamentally misdesigned. The Bank of England, which had no real responsibility for the Stock Exchange's settlement system, but which then as now regarded itself as basically carrying the can for anything that went wrong in the City, decided that if you want something done properly, you do it yourself. So it took a couple of dozen high-flying civil servants and told them to make a world-class IT company.

It's interesting to see how public sector attitudes to technology have changed over the years. Back in the 1980s, the BBC thought it was the most natural thing on earth that they should be setting the specifications for a mass market microcomputer, and the middle class of the UK agreed with them, in large numbers (at my school at least, owning a ZX Spectrum was quite controversial, as it was widely believed that mere private enterprise could never produce a computer as good as the British Broadcasting Corporation. This was rather compensated for by the fact that the Spectrum had good games, whereas the BBC Micro games all consisted of the letter Z chasing a letter H through a maze made of the letter M). On the other hand today, I don't think there's a single major arm of state that hasn't decided to outsource its important IT functions. It is not at all obvious that this loss of confidence has resulted in either better or cheaper technology systems.

Anyway, back in 1993 the public sector still had some vestigial testicles, and so CrestCo was created. And it was a fantastic success; delivered on time and on budget, for quite a long while Crest was widely regarded as the best securities processing system in the world. Even today, the Bank of England are regarded as the go-to guys in most of the central banking world for payments systems design.

There is, perhaps unfortunately, no official history of Crest - the nearest that I am aware of is this 1996 article on its regulatory organisation. So it's difficult to see whether there was some specific thing that they did which made the project work. Iain Saville, the chief executive of CrestCo, certainly did a very fine job in managing the project (and whoever ends up taking responsibility for the NHS mess should certainly look up his number), but the recent demise of Kinnect shows that he didn't have a silver bullet solution for these things, or at least not one that worked sufficiently well against the general tendency of the Lloyd's market to make anyone question their will to live.

I think, though, that the most important reason why Saville's team succeeded in producing Crest on time and on budget was that the Bank of England demanded - and had the clout to ensure that it got - a stable environment to build in. Between 1993 and 1996 there was a constant regulatory and business environment, as new initiatives for changing the way the Stock Exchange did business were suspended pending the deployment of Crest. No "reforms", no "initiatives", no "far reaching developments", nothing. Unlike the Taurus team, the Crest team were able to write code to solve the problems in front of them, without someone moving the goalposts every five minutes.

This would seem like an unbelievably obvious, basic rule of good practice; that you can have major operational projects or major structural changes, but not both at once. It is, in fact, one of the big principles that they teach you in business school. But in the British public sector, this principle appears to be treated with the most monumental and catastrophic contempt. There was simply no chance that the NHS IT project (or the various Home Office projects, or the various education projects) was going to succeed; failure was written into the specification by the fact that the government chose to ignore the existence of the projects when deciding to have a dozen or more attempts at "radical change".

In general, there is too much radical reform. There appears to be a nasty tendency in modern managerial politics to overestimate the benefits of structural reorganisations, and to underestimate the dead weight costs of shifting everything around so often. Not only is the cost of the actual "reforms" themselves often very material, but the effect on morale and performance of the constant environment of "Until Then, Business As Usual" is huge. Even if it were the case that all the problems of the NHS could be solved if we only had exactly the right kind of bogus, playing-at-shops internal market arrangements, the short term costs have to be taken into account, in a way in which the Blair government has systematically failed to do.

We've now reached a critical point in terms of the capability of the systems of the NHS to do their job - the Home Office is now well beyond this critical point and admittedly so. Whatever the benefits of the next round of reforms, they cannot possibly justify the costs of the logistical meltdown they are quite likely to cause.

So it is worrying in the extreme that the one issue on which Gordon Brown appears to agree with Tony Blair is that constant reform is still a priority across the public services. He ought to have declared a moratorium on major administrative change for the remainder of the parliament - anyone who knows a teacher, doctor, nurse or local government officer knows that such a moratorium would have been cheered to the rafters. My guess is that it would be popular too. New Labour sold itself to Middle England on the basis of competence in administration, which is not the same thing as dynamic radical change and is often inimical to it. And in any case, whatever one thinks about the current state of the NHS, it is hard to argue that there is any structural change which is so urgent that it can't wait three years in order to let the systems catch up.

The phrase "The status quo is no longer an option" is reliably the leper's bell of the modern managerial idiot. It is almost always wrong. Like Status Quo, the status quo is often vastly underrated simply because it is unfashionable. The great thing about the status quo is that it is not any worse than the status quo. Surprisingly few proposals for "radical and far reaching reform" can actually beat this standard. If any New Labour leadership candidates are looking for a Big Idea to carry the mission forward, could I suggest "stasis"?

Most viewed

Most viewed