There were plenty of reactions to the government’s infrastructure announcement a few weeks ago which saw them fund a bunch of big roading projects. One of those was from former Transport and Finance minister Steven Joyce. Apart from the general politicking of it, this bit stood out to me.
Now we have more general acceptance that new multi-lane highways should be built, this is a great time to ask ourselves how much needs to be built.
What is the correct size of our expressway network? What should our goals be now that the first roads of national significance are nearly finished?
There’s still a lot to be done. I can’t see Australia for example accepting the lack of four lanes between cities like Tauranga and Hamilton, or Auckland and Whangarei. Or four lane highways petering out a few kilometres outside cities the size of Christchurch.
And I doubt most Kiwis living near those roads accept it either. Meanwhile as a result of Left-wing activism Wellington city remains a permanent traffic jam.
It is true that a lot of our rural highways don’t need the four-laning treatment. There is no point four-laning the Desert Road or State Highway One through Southland because there isn’t the traffic to justify it. A number of well-placed passing lanes do the trick in most places.
However around six per cent of our state highways are high-use.
They account for more than a third of all kilometres travelled on the highway network. Upgrading them is about safety as well as reasonable travel times. Stressed and impatient drivers making rash and risky decisions are an under-rated contributor to New Zealand’s road toll.
With all that in mind, here is a starter for ten on what a new state highway building plan for New Zealand could look like: three networks of modern four lane highways based through and around our three biggest cities of Auckland, Wellington and Christchurch.
The Northern Expressway network would safely and efficiently link Whangarei, Auckland, Hamilton, Tauranga and Rotorua. The Central Expressway network would do the same for Wellington, Hutt Valley, Levin, and Palmerston North, on towards Whanganui and over the hill to Hawkes Bay.
The Southern network would radiate out from Christchurch, north to around Amberley, south to Ashburton and on towards Timaru, and inland towards the Alps.
A decent chunk of each is now already built. Completed over say a twenty year period the three networks would provide safe, reliable, stress-free travelling of a standard that is taken now as a given in the rest of the developed world.
Firstly there are plenty of examples in Australia of cities similar sizes to Hamilton and Tauranga that are not connected by expressways. For example Ballarat (105k) and Geelong (268k) are the same distance apart as Hamilton and Tauranga and the road between them is about the same quality as most of our state highways. In Queensland, Cairns (181k) and Townsville (153k) are a bit further apart at about 300km but similarly have just a two-lane highway between them.
Joyce is correct that that many of our state highways simply don’t have the traffic volumes to justify upgrading to expressway standard. Where I’m sure we differ is the amount of roads that fall into that category. Yes, a third of all kilometres travelled happen on high-use roads but that third can be made up from just the state highways within Auckland, Wellington and Christchurch. As I’ll show later in the post, as soon as you get outside of those areas, volumes drop dramatically.
Perhaps what I thought was most interesting was his suggestions of the expressway networks. Clearly this thought process was been behind much of the previous governments and Nationals current transport policy and while it might appeal to some on paper, that changes if you take more than a superficial look at them. In each of these three networks the red represents expressway/motorway segments that already exist, yellow are projects already underway or announced by the government last month and green are hypothetical projects mentioned by Joyce above or were in Nationals transport policy at the last election.
- Northern Expressway network – Within Auckland I’ve included in here the East West Link and a bypass of Huapai (this hasn’t been on any list but is in the Supporting Growth work). Outside of Auckland also includes an extension of SH1 4-laning to Taupo as well as SH2 from Omokoroa to past Katikati, both of which were in Nationals last election policy. In total these additions are about 330km of new expressway.
- Central Network – I’ve made some assumptions on just where these would go but they amount to over 275km of expressway.
- Southern Network – I’ve assumed by inland towards the Alps he means along SH73 towards Darfield. To complete this network would require an additional 200km of expressway.
How much would it cost?
All up that’s over 800km of new expressway and it’s not going to come cheap. To get an indication of just what it might cost I took a look at the projects just announced and many similar types of projects completed over the last decade or so to provide a reference. As part of this I excluded projects like the widening of existing motorways but I have included a number of 2-lane highway realignments, such as the recently completed Matakohe Bridges as they’ve often involved decent scale earthworks and bridges.
In all cases I looked at a cost per km to provide a better comparison and I adjusted them for inflation. As you can see below, there’s a wide variety of costs as each project will have different amounts of earthworks and structures needed. As such the price range for expressway grade roads seems to vary from about $15 million per km to about $80 million per km.
For the hypothetical network above, in cases where we know the estimated costs, such as from a recent business case, I’ve used those figures, otherwise I’ve used a figure of $30 million per km. In total these motorways/expressways would cost the country close to $27 billion to build on top of existing commitments such as the over $4 billion recently announced. Even spread out evenly over 20 years, as Joyce suggests, it would leave little or no money for any other, likely more valuable, project.
So they’re expensive but do they make sense?
When should we build them?
One of the projects we have a good idea of costs about is Warkworth to Wellsford. The project isn’t funded but recently the NZTA announced they were going to be lodging resource consent for it next month – probably a part of refilling the bucket of ‘shovel ready’ road projects for the next time the government comes knocking for projects to quickly fund. As part of this, they’ve also released the business case for the project.
The road will be 26km and has an estimated cost range of $1.7 to $2.1 billion, meaning it has an average cost of $65-85 million per km. This is much higher than many other four-lane highway projects and one of the reasons for this is likely to be that the project includes an 850m tunnel which they say is to avoid the need for a 150m high cut through the hillside. For a quick comparison, the Johnsons Hills tunnels just south of Puhoi are 340m, the Victoria Park tunnel is 440m and Waterview is 2.4km in length.
However while there are benefits to the project, they’re not that large as the traffic volumes are relatively low (12-15k vehicles per day) and over it’s length it only saves about 3 minutes of travel time. This means the Benefit Cost Ratio (BCR) for the project is just 0.7, meaning it returns only 70c for every dollar it costs to build.
But what’s particularly interesting about the business case is it includes a set of project triggers for when it might be needed. These are:
A mix of criteria is proposed to the trigger the implementation for the Project, being at least two of the following criteria:
- DSI savings forecast from Dome Valley safety improvements not achieved within 3 years
- A 30% increase in total number of closure hours per annum from 2018 levels
- Forecast traffic volumes are predicted to exceed 25,000 AADT
It’s the last one that is most interesting as it provides a useful reference point for other potential state highway projects. For reference, most sections of the Waikato expressway do about 25-30k vehicles per day.
State Highway Traffic Volumes
Finally, something I was actually working on from before the announcement. The graphs below show average daily traffic volumes on a number of state highways. I’ve included a line showing what 25k per day looks like. I’ve also kept the volumes at the same scale
A few things that these highlight.
- Auckland is in a league of its own. Even traffic volumes on the busiest section of motorway in Wellington or Christchurch doesn’t come close to the amount of traffic on most of Auckland’s motorways.
- The busiest section of motorway in the county is between Khyber Pass and Gillies Ave with just over 200k vehicles per day.
- Even existing sections of the Waikato Expressways only carry around 25k vehicles per day.
- Apart from a few urban areas, there’s almost no-where where volumes are close to that 25k per day threshold.
- The first graph really highlights how long SH1 is north of Auckland
The low volumes on most of our state highway network combined with the high costs of upgrading to expressway standards suggests that likely the best course of action are more projects such as those 2-lane realignments in the cost graph above. Combine those with safety upgrades such as median and side barriers, additional passing lanes and we can probably get most of the benefits of an expressway for a fraction of the cost.