How the NHS lost twelve billion: “procurement” is such a dirty word

“Mr Jenkins, this chap says he’s from KPMG. He’s redrafted the PFI contracts for you.” “Ah, brilliant! Send him in, Charlie, I like the man’s style. He’s a snappy dresser, and such great taste in music.” “Are you sure, sir? He seems a bit like a lazy and over-exaggerated visual metaphor.”

Like an asthmatic platypus making a doomed escape attempt from a collapsed toffee apple factory, the botched NHS computerisation scheme has finally wheezed its way to a sticky end. Admittedly, cutting up its bloated carcass reveals a few still-useful organs, making it not-quite-as-massive a muck-up as everyone seems to be suggesting: the central patient records database works, a bit, though it’s only mildly necessary at best, and the digital X-ray sharing wotsit seems to be mostly running well. So that’s nice. Still – for something costing twelve billion pounds (coincidentally, the same amount as that tricky deficit thing that we keep trying to sort out by firing police and gutting charities) you’d hope for a bit more. I mean, according to the sums I’ve just scrawled on the back of my hand, that’s enough money to buy 4792,452,830 boxes of Coco Pops – enough for every man, woman and child in Britain to eat a delicious chocolatey breakfast every day for three and a half years.

(Admittedly, my methodology is hugely flawed here as it assumes everyone sticks to the recommended portion size – a stingy 35 grams, barely enough to cover the bottom of the bowl.)

So something that lets doctors send each other photos of your ribs doesn’t really seem worth it – they could have just got a Flikr account and spent the rest of the day figuring out how many swimming pools they could fill with all those Coco Pops (this is actually quite a tricky question). What happened, then? How did such a simple idea – a better computer system that’s standard across the NHS – grow into an enormous saggy monster that failed miserably in almost every single one of its aims, toppling soggily to the ground like a tower block of wet tofu? The full answers are still hidden away somewhere, but nevertheless  some brave explanations have emerged blinking into the light of day, stumbling cautiously out from beneath the massive rock of turds. They go like this:

The contracts were signed before anyone really knew what they wanted. It turns out that it’s a bit difficult to figure out what’s best for every single one of the UK’s 353 NHS hospitals, and apply it universally. When it came down to actually sticking the kit in the hospitals, everyone changed their mind faster than Gaga changes spangly kickers and everything was left going round and round in endless design meetings until they all got dizzy and had to go home.

The contracts were badly thought-out and poorly assigned. This one’s a bit obvious really, but the point still stands – at the beginning, policy bods crowed about how these contracts would punish any supplier who failed to deliver what they’d promised. But when the suppliers all started to fall away like scabs from an abscess, what happened? They, er, got away with it. Accenture could have been charged a billion pounds for walking away from its £2bn contract; instead it got away with a relatively paltry £63 million. Another main supplier, iSoft – who were tasked with writing the software for these systems – went bum-over-nostrils and had to be bailed out by the government to the tune of £82 million; four of its ex-directors ending up being charged over certain “irregularities” in the iSoft accounts. Whoops!

The tendering process excluded small companies. Amidst all the flap, this is the big solution to the NHS’s computer woes: letting individual hospitals sort their own IT out with small, local projects that can be easily managed. Health Secretary Andrew Lansley went so far as to call this “an innovative new system driven by local decision-making” – the very fact that this idea should be considered innovative just goes to show how squiffy the whole tendering process had become, skewed so far towards massive globocorps like Fujitsu and Accenture that it’s a wonder it just fall over into their waiting arms. Let’s use an example of standards-led design – the MP3, say. It shouldn’t have been too hard for the NHS to get a small technical team to decide the standards for coding up patient files, like the bunch of sexy Germans who came up with the rules for making MP3s. Then they could have left it up to the individual trusts themselves to figure out what sort of system they wanted – in the same way the German team let other people knock together stuff like iTunes or Windows Media Player or whatever, safe in the knowledge that MP3s would always be compatible between them. OK, this approach isn’t perfect – it doesn’t let the NHS as a whole use its freakishly huge purchasing power to get the cheapest deal. But it also would have meant that everyone didn’t get stuck trying to make a one-size-fits-entire-country bonkers electrobox.

Piers Morgan did it. Well, probably. He is a colossal dickhead, after all.

Brandenburg, Grill and Popp. They sound like they might be mascots for Aldi’s own-brand knock-off equivalent to Rice Krispies, but actually these guys did a great job on MP3 so I can’t really bring myself to make much fun of them. Good on you, lads!

This affair might have been relatively forgivable – it was a hugely ambitious project after all – were it not for the fact that this sort of thing happens all the time. Usually the culprit is PFI, an odd financial dodge beloved by successive governments of all party colours, which has private companies own things like hospitals and fire engines then rent them back to the state in a sort of buy-now-pay-more-later scheme: the Public Accounts Committee say it “provides a better deal for the private sector than for the taxpayer,” the head of Unison reckons it’s like paying for a mortgage on your credit card, and Private Eye has an entertaining examination of the whole sordid mess here. PFI leaves the government stuck in contracts lasting twenty or thirty years – it’s got a lot of hospitals built, and some of them have worked out fairly well, but they’re expensive to keep going, and there have been few spectacular implosions along the way.

There’s the amazing story of FiReControl, whose oddly capitalised name makes it look a bit like a Japanese kids’ merchandising phenomenon. It was actually a wheeze first coughed up by John Prescott  to fiddle with the 999 calls system, against the wishes of just about everybody else. In full public view, it went massively wrong on a scale not often seen outside the Katona family, costing nearly £500 million in the process and leaving us forced to keep paying fifty grand a week to rent empty buildings from the PFI contractor. Then there’s Metronet, the PFI consortium that was supposed to run the London Underground and instead just wazzed it all up the wall, collapsing under the weight of its own ineptitude and having to be bought out again by the government. £410 million for that one, which might explain why a single ticket in central London costs four pounds while you can go anywhere you like on the Beijing metro – air-conditioned! – for 20p.

Even the PFI schemes that work out still don’t provide as much value as they should: one of the key assumptions underpinning the calculations behind PFI is that the companies involved will pay tax on the profits they make, thus giving the Treasury a refund of sorts that makes PFI cheaper. Problem is, the rules guiding these sums – the so-called “Green Book” – were written by the consultancy firm KPMG, who also have a nice line in advising PFI companies on how to avoid that same tax by registering their businesses in Guernsey. It’s a nice trick for KPMG, but a kick in the nuts for the public purse. Which, er, is apparently the sort of purse that has testes. Best not dwell on that.

PFI contracts are often hastily-drawn, short-sighted monstrosities – but they’re glitteringly faultless in comparison to the overpriced and badly-tendered reams of bogroll that come streaming out of the MoD printers.  The kind of contracts that spend £500m upgrading helicopters, and then downgrading them again while they sit unflown in a hangar for eleven years. Or hiring a consultancy firm on a single-tender contract worth a cool twelve mill, which wasn’t advertised or opened up to any other bidders. Or allowing employees who work on the contracts to leave and take up a post at the very same company that was just awarded the job – a policy which can backfire in a seriously bogus way, man.

Basically, the government is rubbish at buying stuff. It doesn’t shop around, or think about what gives it the best value. It gets everything on credit, from its mates who assure it that they’re all perfectly legit, and ends up losing billions which it then has to claw back from teacher’s pensions and the dying: groups which, unlike massive arms companies or consultancy peddlers, aren’t in the habit of regularly taking top civil servants out to lavish dinners. I don’t mean to come across as some raving hippy who snorts wheatgrass and thinks shoes are unethical , but still – when the government is spurting cash down the U-bend like it’s the morning after they’ve just gone for a night out on the poppers to celebrate winning  the “Let’s See Who Can Spend a Whole Week Eating Nothing But Vindaloo and Coins Contest,” you’ve got to ask: would it be too much for someone to learn how to write a decent contract? And if not, can I still have my 42 months worth of free cereal?