The City Rail Link is, undoubtedly, Auckland’s most important transport project – with this fact being highlighted in the Auckland Plan, which places the project as Auckland’s top priority. However, it is also a very expensive project – but just how expensive? And what is the cost breakdown? And how could costs potentially be trimmed? Those questions are important to explore further – particularly as reports mount about the CRL’s cost going up and up.
One problem that contributes to undermining support for the project is just how little information Auckland Transport share with the public on issues such as this. Really the only information we have to determine the cost of the project, and how it’s made up, is from an Appendix to the 2010 Business Case and then some further details in the 2011 Business Case Review. Here are some of the details from the original business case, which breaks down the costs:
There are a few important points in the table above, which are worth highlighting:
- The physical works of the project cost around $981 million out of the total budget of $2.4 billion. The rest seems to be property costs, contingency, design, client management, contractor’s margin and something called “funding risk”. I’ll call these the “other costs”.
- The $2.4 billion total is actually the 95th percentile estimate (that is, there’s a 5% chance the cost will be higher and a 95% chance the cost will be lower). The “actual” cost (which I presume could be thought of as the most likely cost) is almost bang on $1 billion.
- The stations make up a significant portion of the project’s physical cost: Aotea, K Road and Newton stations have a combined cost of around $455 million (excluding the “other costs”), around 46% of the project’s physical cost. Including all “other costs”, K Road and Newton stations are more than $300 million each.
As well as the “core cost” of the project (which has a closest estimated cost of $2 billion), the original business case also highlighted a number of additional areas of cost that would be required to implement the City Rail Link. These are summarised below:
The $100 million in associated rail infrastructure works includes duplicating the Onehunga Line and grade separating a number of level crossings along the Western Line, as detailed in Appendix F. Whether these should be included in the cost of the CRL, just as whether the additional rolling stock should be included in the project’s cost (you don’t see the additional cost of buses included in business cases for bus projects) is an interesting debate to be had.
Debates between Central Government and Auckland in the Business Case Review have been very well covered on this blog over the past year, but what was interesting is that the parties generally agreed in their assessment of the project’s cost. This is outlined in the Workstream 6 document:
Our reviewers raised more than two dozen points ranging from major concept suggestions to relatively minor cost queries and opinions, which were referred to the APB&B Study team and Auckland Transport for their views. Most of these suggestions were countered, giving confidence that the original concept and costs had been considered in a robust manner. Some of our suggestions will be carried forward to the detailed design stage, when that is approved.
Both our reviewers were very interested in the benefits to the concept design and construction methods that might be gained through having contractors input to the project. Contractors’ knowledge of techniques and technology could potentially be used to save time and money by optimal design, staging, and construction. Our reviewers are of the opinion that it is conceivable that, given contractor input, it may be possible to reduce the costs if other innovative ways of working could be achieved, and this forms one of our conclusions.
They both agreed with the peer reviewers that the project could be carried out for the estimated cost (excluding property costs) of between $1.9b and $2.2b. Thus they confirm that the estimated construction costs are realistic.
In summary, it seems that the NZTA reviews found the cost estimated in the original business case to be realistic, and if anything their feedback suggested that savings were possible through an innovative contracting process.
The significant number of peer reviews the costing has gone through (APB&B themselves had the costs reviewed by three different external agencies) suggests that these numbers are pretty robust. Which is why it is rather surprising to see higher numbers being bandied about in more recent times.
Coming to our last question, how could the project’s cost be trimmed a bit, we learned from commenter “Greg N” (who himself learned from Auckland Transport’s Stephen Rainbow) last month that various options are being looked at on that very issue: in particular delaying the construction of some of the stations (presumably either K Road or Newton station or both, as they are the most expensive and each are likely to generate much less patronage than Aotea Station) and/or not building the eastern link between Newton and Grafton.
The eastern link, K Road and Newton stations are three very expensive parts of the project, if we recall the table earlier on in this post. If we assume that adding in the “other costs” (design, contractor profit, contingency etc.) doubles the actual cost from its pure construction number, we get the following numbers for those three aspects of the project:
- K Road station: $319.6 million
- Newton station: $315.2 million
- Eastern Link: $180.8 million
A grand total of $815.6 million in those three pieces of infrastructure which seem to be ‘up for debate’ over whether they will actually happen (when the project is initially constructed). Obviously there are good reasons to build the two stations and it seems like there is an argument for the Eastern Link (plus probably a pretty significant infrastructure requirement if you don’t build it as you’ll need to turn a heap of trains around somewhere on the Western Line), so there’s a debate to be had around whether they should be in or out, but it’s fair to say that the cost saving potential is rather significant. Of course ‘future proofing’ for the two stations will incur some cost, which is another matter to consider.
But to come back to the question in the title of this post – what does the CRL cost – is seems the answer is fairly complicated and a bit of a movable feast. But what do we know?
- We know is that the core cost of the project has been assessed over and over again (with three stations and an eastern link) at around $2 billion.
- We know that a significant chunk of this could be saved (say $500 million and I feel like I’m being conservative with that) by delaying the stations and perhaps not building the Eastern Link.
- We know that some extra trains will be needed, and some infrastructure works around the rest of the rail network, though it’s debatable whether they should be included as part of the project.
All up, it seems like there’s a pretty damn good chance the project could end up being under $2 billion – including the additional trains and additional infrastructure needed, if the stations are delayed. If this is the case, then it’d be great to have Auckland Transport start telling the world about this very soon before the project loses even more public support. It would also be good if those that do oppose the project actually came up with some ways that we will be able to move people around the region in 30 years time when there are potentially up to 1 million extra people living in it.
Processing...
It seems particularly absurd to include additional trains required in 2030… I have never seen any motorway costing including all the new cars and trucks that we will all end up having to buy to use it. Especially when other alternatives are not invested in.
The motorway system has always been sold to us in little chunks that look more palatable yet it still adds up to being the single most expensive way to provide movement in a city. Never included in any calculations are the enormous costs of vehicle purchase, maintenance, and especially parking- a huge burden on land use and building development. Waterview/ Northwestern will cost more than 3 billion excluding all these addition cost burdens on us all, and will simply add to the choke point at the Central Motorway Junction.
The value of the CRL- that it provides the capacity to move the equivalent of ten lanes of new motorway all without demolishing any valuable property nor building new carparking nor burning more imported oil, at around 2billion we can’t afford not to build it.
The benefit cost ratio of the project was calculated over a 30 year period,so it is appropriate to include all capital and operating + maintenance costs expected to be incurred over that period. The additional trains projected to be needed by 2030 would be required to carry the projected patronage demand
I’ve yet to see a BCR for a motorway that included future fuel and vehicle costs over any period….
Another thing I forgot to mention is that we saw a similar process happen with Waterview where the cost ballooned up to about $3b at one stage before coming back down again as the project was refined. We are probably at the same stage with the CRL
Matt, you raised a very good point at the end. The debate should not be about tunnel or no tunnel, it is really about choosing between:
o A tunnel: Everyone can get to the CBD on time in 2040 and the CBD isn’t full of cars or buses.
o More roads into the CBD which will certainly cost more than the tunnel. Cars in 2040 can still expect unreliable journey times into the CBD. Parking will always be expensive. CBD streets will be full of polluting cars.
o Do nothing – we have lots of experience in this. In 2040 nobody can get to the CBD on time, business look at moving into the suburbs, eventually get road congestion in the suburbs as well.
So actually the ~$2billion of the tunnel is quite cheap compared to building more motorways, and even compared to doing nothing, $2b is not much. We are a city of 1.5 million after all. We aren’t Wellington or Christchurch.
Thank you Matt for this post 😀
If my Facebook feed was anything to go by there is a lot of activity and panic inducing stuff coming out of the Centre-Right with costs running away and so on.
Now to save me contradicting myself I have stated in the past the CRL would be at $3.6b and 7 seven years to completion. This was based on a rail fallacy out of the EU (of all places) that any figure a public authority gives for a rail project; times the cost and completion date by 1.5x and you get a closer reflection of costs. The USA especially Hawaii (which I posted here recently) did provide an example of The Rail Fallacy in play (the actual scheme has merit, the cost and completion figures ran away and political selling of the project went sour (hmm sounds familiar to here?). The Manukau Line which is by far a more simple project illustrated the Fallacy with the project 1.5x over budget and I think 18 months late while still 435m short (and I continue to hound Dick Quax over it).
So one could see where I was coming from (and to which I am writing a feature post over the next few days on The Rail Fallacy).
For Matt, Nick, Patrick and others I am explaining my logic here that I am using for the CRL. $3.6b for the cost by the time project starts and most likely be done by PPP with help of the Chinese. Why $3.6b? You can see costs getting ramped up and up and up which sours the debate – sure costs can be drawn back but often too little too late. The Manukau Line and Hawaii is not inspiring confidence which the Centre Right know and are using. Thus if the Mayor said last year $3.6b for starting the project 2018, it would of woken Auckland up all right but we have 6 years for a debate, finance raising and getting the Chinese on-board. If savings come about then a win for Len and a promoter in the debate rather than the opposite seeing now – and I bet my coffee the City would be more in favour for this project.
I have seen the report from Matt and discussed it at length with work mates, the conclusion (after also seeking some external advice) we came at is YES build the CRL in a staged development however we are looking at $3b approx at the moment given history. We also tried to find savings as well and found a few 😀
Look before anyone decides to point cannons my way. I support the CRL I really do and I hope I am proven wrong when the CRL is opened when it does, however (and I having a similar situation with some else over China) history tends to repeat itself and given it here, well yeah…
EDIT: yay an edit and delete button. Me is happy now
So you’ve taken the upper bound cost estimate (already factoring in escalation) and arbitrarily multiplied that by one and a half to give it super heaps more escalation? Sorry there isn’t much logic to that, just opinion.
Well Ben that’s nonsense. There is plenty of contingency in those figures and in fact there is good reason to expect a lower figure though competitive tendering. Inventing cute ways to multiply numbers that already include extraneous extras like track work elsewhere and future rolling stock is just throwing red meat to nay-sayers like Mr Brewer. Like you he can’t fault the need and the value in the project so he has to claim it as unaffordable, why make up your own fantasy numbers just to aid this miserable short term thinking?
Cameron Brewer was the only councilor to win majority support (over 50%) in his ward. Does the rest of the council really represent the views of the people?
Brewer represents the most homogeneous ward is all that tells you. Stand him in South Auckland and see what his majority looks like. Not many live in Remmers without lining up for the Cits ‘n’ Rats world view.
Ben,
As others have pointed out, I think what you are saying is just not right in this case. There was a time in the past when that may have been true. But in the past the quoted cost would have been $1b. Things have only ever cost as much as they cost, it is just that in the past the engineering proffesion used to systematically underestimate costs. This is not the case anymore. Look at NZTA’s recent projects – they are far more accurate these days because people price in risk as a matter of course. As people have pointed out above, the cost of Waterview at $1.4bn is lower than the quoted $1.5b back in 2006.
Matt thanks for this post – it has helped me make sense of the CRL costs. I previously couldnt understand why they were so high.
Over $200m of design and “client management” fees though. Woof.
What about cutting some of that money ($815m), and looking at spending $100m on reboring the Waitakere tunnel and extending the electrification to Huapai and building a small scale stabling facility there. This could extend the catchment, and increase the patronage, also get some of the equipment out of the area to, reducing train congestion. Of course track upgrade maybe required, and some signals, but couldn’t imagine it being a major thing, given it is single track. Keep in mind, it’s mainly a straight section to Huapai, and if the track is upgraded, the line speed could be increased from 70 to 80, 90, 100…
I suppose I’m suggesting here, get the tracks under the CBD, and at least a major stop on the route, with tracks going from the west and into Britomart. Also suggesting get the wires up to Pukekohe, and make this look a bit more serious, and worthy to all of Auckland, not just some.
Spending some extra money on an Avondale to Mt Roskill spur would be a far better plan than a Swanson to Huapai extension used by 100 people a day.
Great post Matt L!
Put all this info into a snappy looking website / microsite with the address crl.org.nz or build-crl.org.nz etc. The site should become the focal point of debate and imformation dissemination / clarification, even after AT finally get around to releasing more info. Your AT Blog on its own is not enough to communicate to the wider AKL public. A hot topic such as the CRL needs to have its own website.
leaving out the 2 stations (but future proofing for them) seems to follow the logic of not including WEBs such as agglomeration in the business case. Less so for the eastern link, but you’ve suggested it’d be the most likely to be kept if one was going to be.
You would still get agglomeration benefits with K Road and Newton station delayed. In fact, if the development that could have happened at all three sites (Aotea and those two stations) was to be concentrated around Aotea station then your agglomeration benefits might be even higher.
The eastern link is critical to the long term functionality of the CRL, and I can’t see how it could be built at a later date if it were left out. To me that suggest they need to keep the link and stage the stations instead.
Is there a breakdown of the station costs?
$319.6M & $315.2M sound like some fairly precise figures which must have a fair bit of detail behind them.
I’d be interested to see why these holes in the ground cost so much.
Compare these costs with other construction projects, e.g. $201M for 4.5km SH20 Mt Roskill extension which included 4 road bridges, a big cut & retaining through Hillsborough, ground dewatering & stabilisation through Keith Hay Park, and cutting through the basalt of Mt Roskill.
The costing should also show the construction method being considered, i.e. dug form above or mined out underground.
If dug from above, what consideration has been given to the use of the space between the station & road above (e.g. retail, theatres, supermarket, carparking, etc) and the development costs?
“It would also be good if those that do oppose the project actually came up with some ways that we will be able to move people around the region in 30 years time when there are potentially up to 1 million extra people living in it.”
I think the plan of those opposed to the project is to not build it, thus guaranteeing that Auckland will not be home to 1 million more people in 30 year’s time, because it will be a living hell of a place to get around, and no businesses, and no migrants in their right minds would want to come here. Problem solved.
How does a single station cost $320 million?
Bloody long waybunderground in the middle of he city. But they won’t, there’s a hell of a lot of contingency etc in there…..
One thing the CRL will do is create new places out of nothing. . I have not heard anyone arguing against the CRL beside cost. This is quite different to roads and even buses which take away resource and impact on others. Often “just” fields ( although our agriculture is one of our biggest earners) but sometimes peoples houses and sometimes splitting communities.