This is a post about the cost overruns at the National Children’s Hospital. This is big news lately, with some cost overruns that appear truly staggering – from an initial approved costing of €570m, to a point where the government have acknowledged that it may cost €2bn or more.
The problem with the question of costs, and overruns, is that fairly subtle and detailed issues can end up costing enormous amounts of money. The quoting of multiple, massive figures in the media just begs any observer to roll their eyes, mutter “typical,” and forget about it. In fact, some areas seem to be very atypical, and some are worryingly commonplace. This is an attempt to unpick the difference.
Going through cost overruns is difficult, because there are different kinds of extra cost, and in this instance we’ve only got incomplete information. Therefore, it’s hard to tell them apart. Some of them are, frankly, inexplicable; they could be due to some terrible mistakes by people in charge, or vacillating by the client, or some other factor we haven’t been told about yet. Some of them appear to be down to deliberate omissions, i.e. things that weren’t included in the initial costs but were always known about and aren’t a surprise. And others are the sort of inevitable overrun that will always occur on a project of this size, no matter who’s building it-
What do you mean, inevitable? Sounds like you’re making excuses.
Not really. Some increases are inevitable. Inflation, for example. You do a costing at the very start of a job. If it’s of even a vaguely large size, it will take a year to get from there to the point where you’re asking the contractor to price it – and prices will have gone up in the meantime. In construction, standing still costs money. You can, if you’re clever, factor inflation in from the start – but you don’t really know how bad it will be, so you’re guessing.
Between start and finish, there are many other ways that a job can go over budget – or at least, that the budget has to increase. The faster you want to go, the more likely this is to happen.
For starters, the design brief will change. If you’re doing a large, complicated building, you might have to consult with hundreds of individuals to figure out exactly what they want. No single person knows exactly what all the surgeons / radiologists / occupational therapists / insert-specialism-here need and besides, the criteria are always changing as their professions evolve. You could compile all that information before you start, but it will take a long time (see the bit above about standing still costing money). Besides, you still have the human factor of someone coming back with “Sorry, I forgot to say, I also need…” – so you start, using the best information you have to allow enough money, and fine tune as you go.
And there’s mistakes. When that initial cost estimate is made by the Quantity Surveyor, they don’t know all the information – they can’t know all the information until the building is finished. So they make assumptions. So does everybody else – the architects make assumptions about how the design will work, the engineers make assumptions about the loadings, the mechanical and electrical engineers about the level of services needed. Those assumption should be professionally made and broadly reasonable, but they won’t all be right.
Well, I don’t hear about private sector jobs going this far over budget…
Look, don’t get me wrong. Something has gone really horribly wrong with the National Children’s Hospital and we’ll get to that in due course. But there’s a few points here.
First of all, private sector jobs do go over budget, too. You don’t always hear about it, because they don’t announce at the start what the building is going to cost. But google some of the stories about Tottenham Hotspur’s new stadium and you’ll see that any project can get into a mess.
There’s also a couple of issues that apply to the public sector and are unavoidable. Firstly, public sector jobs tend to be more complex than private sector ones; and also, because the private sector jobs have a fixed budget, they will tend to adjust what they’re building rather than pay more. If a private sector job is 50% over budget, they’ll cut the building back. With something like the National Children’s Hospital, you can’t reduce its size or compromise its finishes or leave out a couple of departments. You have to pay more.
OK. But why are all the different prices being reported on the NCH? Why does nobody seem to know the actual cost?
Part of what makes this difficult to unpick is that there’s a number of different figures floating around and being reported. There’s the initial government costings; the initial tender price; a revised tender price; and the estimated final cost, allowing for future overruns. In addition, there are two variations on the figures; a cost for the “core” scheme, i.e. just the hospital, and a cost for the overall scheme which includes satellite facilities at Tallaght and Connolly hospitals. Plus, those figures are sometimes quoted with VAT, and without.
Essentially, the cost has gone up at every stage. The cost for the overall scheme appears to have gone up by (roughly) a further €450m, and the tender process alone inflated the price by €260m.
Hang on, isn’t the price meant to be fixed at tender? The contractor quotes a price to get the building built?
So why are the costs still going up?
OK, we’ll start from the start; most people struggle to see why, if you are hiring a contractor to do a fixed-price job for an agreed fee, costs should overrun at all. That, surely, is what a “fixed-price contract” means. After all, if you order some clothes from Littlewoods, that’s a contract – and it’s not like they can show up at the door and say “actually it’s forty quid more than we promised, bad luck!”
The costs of construction contracts change because they’re designed to change. Having a construction contract with a completely fixed, no-extras-at-all cost is technically possible, but it would be completely uneconomic to do so. Think what would be involved and this becomes clear. The design team already produces heavily detailed drawings, but drawing and itemising every single thing on a reasonably complex building would go from (say) 4 months to something like 9. The contractor has to take the risk on anything going wrong, even things beyond their control. You have to produce an incredibly detailed survey so that there are absolutely no overruns whatsoever. And every professional involved – dozens of people – has to agree to make no mistakes whatsoever, not even the tiniest error, or to cover the cost if they do.
You can ask people to do this. But you’re asking them to take a huge risk, and they will charge you a small fortune to do it. Nobody – not the private sector, not the public sector, nobody – builds things this way, because it doesn’t work. Instead, the professionals involved warrant to do a professional, competent, skillful job and your design team work bloody hard to limit your exposure to extra costs. You allow contingencies. And you have contracts that allow costs to change to cover unforeseen extras. If you didn’t do this, nothing would ever get built at all.
OK, but is that model – design it, get a price, and accept some cost overruns – the only way?
No. One of the most popular is design-build. This is where you get the contractor to design and build the building – he hires the architect, not you – based on a fixed cost. Often, these do come in on budget (although that’s often, not always). They’re also generally agreed to be more expensive, and – crucially – the quality of the detail and finishes tends not to be as good. So you get an inferior building for more money, but on the plus side you do have cost certainty. That’s the theory, but having worked on many design-builds, I don’t believe it’s a good way of procuring something like a hospital, where high building standards are critical.
The National Children’s Hospital used a hybrid of the two, called a 2-Stage tender. This is often used for large, complex jobs. The design team produce an outline design – maybe the level you might see for planning permission, with a bit more information added – and an approximate bill of quantities is produced. After this there’s a Stage 1 Tender, where several firms bid based on this outline information. The lowest is then selected as the preferred bidder, and can give feedback as a more detailed design is developed. Once the detail design is finished, the contractor gives a revised cost – the Stage 2 tender – based on the completed design and the rates they previously submitted. At this point, there’s an opt-out; the client can decide that the costs are too high, and stop the project.
There are two big benefits here, in theory. First, it’s quicker; and second, you get some contractor expertise during the later design stages.
How well did this 2-Stage process work?
That depends how you look at it. The initial costing from the government for the “core” hospital was €650m. The lowest tender, from BAM, came in at €723m (€637m + VAT). Given that over 2 years of high construction inflation had elapsed between the costing and the tender, that’s not really too bad. The government intended to find savings during Stage 2 which would get the project back on budget, but they only found about €20m – and ran into all sorts of overruns.
The build time was extended by 9 months, for reasons as yet unstated; this added €102m. Engagement with the clinicians, plus the detailed design of all the equipment they needed, went way over budget; it came in €130m more expensive. Throw in some other overruns (remember that story about the sprinklers?) and you have a tender figure of €1bn (€890m + VAT). That’s an overrun of €350m from the initial €650m costing on the core hospital facility.
That’s a bit bloody much, isn’t it? And we taxpayers have to fork out the extra.
Yes, it is a bit bloody much, and throws up more questions than answers. Why was the build time extended by 9 months? And how was the equipment estimate so far out? Most importantly, what works are not going to go ahead, in order that the Children’s Hospital can proceed?
All that said, it’s not necessarily “costing you money,” as such. It doesn’t automatically mean that the public are getting poor value for money; it could mean that the €650m estimate was far too low to begin with. What’s more disturbing is the possibility that other projects could be denied vital funding because the government got their sums wrong in the first place.
What about these other figures? I’ve read €1.4bn, €1.7bn, €2bn…
This is where it gets confusing. These are costs for the overall project, including the satellite facilities, including or excluding an additional IT-based package.
The overall project, including the St James’s hospital and the satellite facilities, was costed at €983m by the government in 2017. This excluded a further €291m allowed for works on IT and various other “extras,” which is being carried out by a separate government body and so was costed separately.
The €983m figure is now being quoted at €1.433bn, so a €450m overrun. The IT isn’t doing too badly – it’s gone from €291m to €300m, giving a grand total of €1.733bn. The “over €2bn” figures seem to be political guesswork, based on various other costs that various health bodies might incur moving into the building, but it’s impossible to tell how real these are.
So what is the overrun?
The work on the St James’ site has massively overrun, going from a €650m initial estimate to just over €1bn. We know roughly what has caused these increases, but don’t know why. It would appear that the initial €650m estimate was simply too low to begin with.
The satellite facilities have gone up €100m, from an initial cost of about €330m. We’ve got very little detail on how or why this has happened.
And finally the package of “extras” has gone from €291m to €300m, a pretty modest increase.
This gives a total of €460m. About €55m of this figure is VAT, and therefore not massively important – after all, VAT is a tax, so ultimately it goes back to the government – it’s simply money that the government is paying to itself.
This is still way too much and the public deserve answers. Most of it seems to be costs going up at the detail design stage as the players involved realised just how complicated the hospital was going to be.
Are contractors lining their pockets? And what’s all this about low-balling?
Leo Varadkar referred to low-balling yesterday; essentially that contractors artificially lower their prices at tender stage, and then pile the costs back on after they’ve been awarded the job.
Well, it happens. Contractors don’t do this for fun; contracting is a very expensive, risky business and they do it to make money. The big players are experts at getting jobs, and making money from jobs. You do the first by keeping your tender prices as low as possible. You do the second by pushing your prices after tender as high as you can, in as much as the contract allows you.
Ireland’s entire system of tendering encourages contractors to find weaknesses in the tender drawings and specifications or in the building contract, omit whatever they can, and lower their price as much as possible at tender – and hence be the lowest tenderer and be awarded the job. Afterwards, the cost of those omissions will be loaded back onto the project as an extra. This is a recurring hazard of any architectural job, one that means Ireland habitually produces poor quality buildings, and could do with serious examination.
There’s a lot more I could say about this, about how the government contracts were replaced by Public Works Contracts in the noughties, and how these sort of “attacks” on the contract became more and more common once the government had (needlessly) replaced a 25 year-old court-tested document with a new, untested one that had a great many issues. I’ve got anecdotes about millions of euro worth of work being omitted from tenders, because a sentence in the contract isn’t phrased very clearly. I can talk about how unreasonable clauses in the government contracts have directly put people out of business. Maybe I’ll get to write all that fun-filled stuff in the future.
But: at the moment, it doesn’t seem to have been the big problem. The Stage1-to-Stage2 cost increase has €22m in extra payments to BAM, which is a substantial figure but – let’s face it – the government wishes that was the problem.
(It may transpire that some of the other additional costs come from BAM leaving out things that other contractors included, and then the issue of low-balling tenders becomes more significant. At the moment, whatever Varadkar says, I can’t find any evidence that anything of the kind has happened.)
The truth is that, much as people like Steven Donnelly have got good soundbites out of €1.5bn increases, the overspend is nowhere near that much. A €1.28bn project has gone to €1.73bn, and counting. A good chunk of that is down to things like inflation, and to some extent it’s because the project was probably underpriced to begin with.
You mean it’s a non-story?
No, not at all. Apart from the costing being way out, another aspect of this is how hard it is to get information about the costing at all. There’s a difference between openness on paper – putting facts in the public domain – and really making figures available. It brings to mind The Hitch Hiker’s Guide To The Galaxy: “The plans were on display in a locked filing cabinet in a disused basement with a sign on the door saying Beware of the Leopard.”
The government could have widely circulated a grand, all-inclusive figure of what they thought this would cost from the start, and asked us to judge them on this commitment. Instead, they appear to have gone out of their way to split everything up and make the total expenditure unclear, occasionally releasing figures at Public Accounts Committees and hoping nobody noticed. Now we have junior ministers in front of Dáil committees, with opposition parties seem to be going out of their way to shout at as many numbers as they can.
The responsibility of politician is to explain to us, the people, what’s going on in the business of government. What we have here is a clear pattern of them trying to avoid giving us that information. That, in my view, is more damning than any multiples of €1bn that get shouted from the rooftops by TDs looking for a headline – and very few of our leaders seem to care about fixing that at all.