This post was first published by Patrick in January 2016.
There are many reasons to be concerned about the plan to add more road lanes across Auckland’s Waitemata Harbour: from the extreme cost of building such big tunnels and interchanges [$5-$6 billion and four times as much as just building rail tunnels], to the undesirable flooding of city streets and North Shore local roads with even more cars, to the increase in air pollution and carbon emission this will create, the loss of valuable city land to expanded on and off ramps and parking structures, to the impact on the harbour of exhaust stacks and a supersized motorway on the Shore, to the pressure this will put on the rest of the motorway system particularly through the narrow throat of Spaghetti Junction. It is both the most expensive and least efficient way to add capacity across this route, and if resilience is the aim then the double-down on reliance the motorway system rather works against this. This one project will simply crowd out any other changes we could make of scale in Auckland or the country for years; yet it changes almost nothing; it simply enables more vehicles to travel across a short point in the middle of the city, yet this is by no means an obviously good thing: The list of unwanted outcomes from the current proposal is so extensive that the benefits had better be so extraordinary and so absolutely certain in order to balance them all.
But perhaps there is no greater reason to not do it than that it simply won’t improve things for drivers.
Really? How can this be? As well the obvious problem with this project that it will add super capacity for a short stretch of the motorway network and therefore just shifts any bottleneck to the next constriction, particularly the extremely difficult to expand CMJ or Spaghetti Junction, there’s also a bigger structural problem with building more roads to fight traffic congestion. It can’t work. We all have experienced being stuck in traffic on a motorway and sat there wishing if only the authorities had just built an extra lane all would be sweet, well it would, wouldn’t it? However the evidence from all round the world shows that while that may help for a little while it never lasts, especially in a thriving city and especially if these extension starve the alternatives of funding, condemning ever more people to vehicle trips on our roads. Soon we’re stuck again wishing for another few billions worth of extra lanes all over again.
Here’s how it works; each new lane or route simply incentivises new vehicle journeys that weren’t made before; a well known phenomenon called induced demand. Road building is also traffic building, the more we invest in roads the more traffic and driving we get, and not just on the new road; everywhere. Traffic congestion is, of course, simply too much traffic, too much driving. Take for example the I-10 in Houston, the Katy Freeway. In that famously auto-dependent city they freely spent Federal money and local taxes disproportionately on just one way to try to beat traffic congestion, the supply side: ever more tarmac [Houstonians can boast the greatest spend per capita on freeways in the US]. The I-10 which began at six to eight lanes has just had its latest ‘upgrade’ to no fewer than 26 lanes! That ought to be more than enough in a flat city with multiple routes and only half the population Los Angeles. So what happened? According to recent analysis it has made driving this route significantly worse.
Traveling out I-10 is now 33% worse – almost 18 more minutes of your time – than it was before we spent $2.8 billion to subsidize land speculation and encourage more driving.
But hang on, those trips must need to be made, right, or people wouldn’t make them. Well in the absence of direct pricing it is hard to know exactly how valuable these new trips are. So first they really ought to price routes like the I-10 properly to reduce unnecessary journeys clogging up the valuable ones, like the truckies and trades [it is partially tolled now]. But the real problem in cities like Houston is the absence of any useful alternatives to driving [an earlier extension of I-10 took out an existing rail line!]. Providing those alternatives is how congestion is best dealt with. Not completely solved of course, that can only happen by collapse of the city economy like in Detroit, and no-one wants that solution. But traffic congestion can be made both manageable and, for many, no longer an issue, by providing them with attractive alternative options. And in turn this frees up the roads sufficiently for those who have to or prefer to drive. Especially when this is done in conjunction with direct price signals- road pricing; tolls or network or cordon charges.
Houston may be forever too far gone down this hopeless road but that doesn’t mean we have to follow it. Here is a description of the same problem in Sydney, with the solution:
Most people will take whichever transport option is fastest. They don’t care about the mode. If public transport is quicker they’ll catch a train or a bus, freeing up road space. If driving is quicker, they’ll jump in their car, adding to road congestion. In this way, public transport speeds determine road speeds. The upshot is that increasing public transport speeds is one of the best options available to governments and communities wanting to reduce road traffic congestion.
This is called the Nash Equilibrium [I would rather say better than faster; there are a number of variables including speed that inform our choices];
This relationship is one of the key mechanisms that make city systems tick. It is basic microeconomics, people shifting between two different options until there is no advantage in shifting and equilibrium is found. We can see this relationship in data sets that make comparisons between international cities. Cities with faster public transport speeds generally have faster road speeds.
Which brings us to the Waitemata Harbour. It currently has 13 general traffic lanes across two bridges, one walking and cycling lane on the upper harbour bridge, and some ferry services generally not competing with these crossings. The Harbour Bridge carries increasing numbers of buses from the hugely successful Northern Busway, the very success of which exactly proves the theory of the equilibrium described by Dr Ziebots above. In the morning peak the buses carry around 40% of the people without even a single dedicated lane on the bridge itself. And it is all the people using the busway that allow the traffic lanes to move at all. In fact NZTA argue that one of the main reasons for building a new crossing is the numbers and the size of the buses now using the current one.
The Upper Harbour Bridge is about become significantly busier because of the multiple billions being spent on the Waterview connection between SH20 and SH16, the widening of SH16, and the bigger interchange between SH81 and SH1 on the Shore. These huge motorway expansions will generate more traffic of course, but also will provide an alternative to driving across the lower Harbour Bridge.
What is missing anywhere between the North Shore and the city is a Rapid Transit alternative to these road lanes. Like Sydney always has had.
It is its [Sydney Harbour Bridge] multi-modality that makes it truly impressive, some 73% of the people entering Sydney on the Bridge from the Shore at this time are doing so on just one of the train lines and one bus lane; a fraction of the width of the whole structure. So not only does it shame our Harbour bridge aesthetically it completely kills it for efficiency too.
Auckland’s bridge was always only ever designed for road traffic, and should be left that way, the clear way forward is to add the missing Rapid Transit route as the next major additional crossing [after adding the SkyPath to the existing bridge].
In 1992 it [Sydney Harbour Bridge] was supplemented by a pair of two lane road tunnels that up the cross harbour tally for this mode to match the number coming over by train [bridge plus tunnels = 12 traffic lanes], but that wasn’t done until the population of the city had hit 3.7 million. The high capacity systems on the bridge saved the people of Sydney and Australia from spending huge sums on additional crossings and delayed the date they were deemed necessary by many decades. But anyway, because the additional crossing is just road lanes it only adds around 10% extra capacity to the bridge. To think that the government here and NZTA are seriously proposing to spend multiple billions in building a third Harbour Crossing in Auckland with the population only at 1.5m, but not only that but they are planning to build more capacity for the least efficient mode; more traffic lanes.
The good people at NZTA of course know this, but we just seem stuck in a bad habit of road building in a similar way as Houston is, because the money for motorway building comes from central government some people believe this makes it free, in a similar way that the highways in the US are largely funded by the Federal government, unlike public transport, which is more locally funded [Known as ‘path dependency’ and is well covered in the academic literature: Imran, Pearce 2014]. This means the pressure to evaluate the effectiveness of motorways over the alternatives is much weaker. Here is a slide from an NZTA presentation proudly proclaiming how much more traffic this massive project will generate:
Of course this growth can be met by a parallel Rapid Transit system instead. The success of the Busway here and the enormous uptake of the recently improved Rail Network show that Aucklanders are the same as city dwellers everywhere and will use good Transit systems when they get the chance. And two much smaller and therefore cheaper train tunnels have much greater capacity than the proposed six traffic tunnels. Twice as much in fact: the equivalent of twelve lanes and without adding a single car to city streets. Furthermore converting the Busway to a rail system, which is entirely possible, and depending on the system may even be quick and easy, means that buses can be completely removed from bridge freeing up more capacity there for general traffic; cars and trucks:
- Removing buses from the existing bridge would free up some capacity. 200 buses per peak hour ~= 1,000 cars ~= 60% capacity of a traffic lane. So a dedicated PT crossing provides car users with an extra lane (once you account for reverse direction). Not huge, but not negligible either.
- Mode shift: by providing a fast and more direct alternative route you will get mode shift, providing more space to the cars that remain. So you have more vehicle capacity and less demand = a real congestion benefit.
So compared to a new road tunnel where both crossings would need to be tolled, and simply generate more competing traffic for drivers through the whole city, the dedicated PT option would seem to be better even for motorists. The better, faster, and more attractive the Rapid Transit route the freer the driving route will remain; with more people choosing the car-free option: The higher the Transit utility; the higher the driving utility.
Of course while a rail crossing will be considerably cheaper to build than a road crossing it still needs a network either side of the harbour to make it useful. Are there good options for this? In fact there are a number of very good options, all with varying advantages and disadvantages that need serious investigation. And it is important to remember by the time this project is being built the public transit networks in Auckland will be considerably more mature. The City Rail Link will have transformed the newly electrified rail network to a central role in the city, it will quickly have doubled from 2015’s 15 million annual trips to 30 million and more. The New Bus Network will be functioning and with the new integrated zonal fare system meaning people will be used to transferring across routes and modes to speed through the city. The increase in bus numbers and population will make driving in the city less functional. There will certainly many tens of thousands more people in the city without their car, many with business or other reasons to travel across to the Shore. And importantly there will almost certainly be a new Light Rail system running from the central isthmus down Queen St and terminating downtown.
The quickest and cheapest to build will probably be to take the city Light Rail system through Wynyard Quarter and across the harbour, as outlined by Matt here. The busway can be most easily converted for this technology, as it is already designed for it. Furthermore being the only rail system that can run on streets it can also most easily include branches to Takapuna and even Milford to the east, and from Onewa up to Glenfield. This also has the advantage of balancing the existing city-side routes, unlocking a downtown terminus, not unlike the CRL does for the rail network.
Higher capacity and with the great advantage of cheaper to run driverless systems are is Light Metro like the massively successful SkyTrain in Vancouver. As described for Auckland here. However like extending our current rail system to the harbour it would require a more expensive city-side tunnel to Aotea Station for connection to city network. We know work has been done to prepare Aotea station for this possibility. Matt has also explored other variations here.
Perhaps the best answer for both the near term and the long term is to build tunnels that can take our new Light Rail vehicles for the years ahead but are also capable of being converted to the higher capacity Light Metro when the demand builds so much to justify the further investment of the city tunnel between Wynyard and Aotea Station. Bearing in mind the LR vehicles AT are planning for are high capacity [450pax ] and they can run in the cross harbour tunnels and the busway at very high frequencies. And that Light Metro systems can use track geometries much closer to LR than can conventional rail systems.
So in summary, the bane of the motorist and the commercial driver, traffic congestion, is best dealt with on the demand-side as well as the supply-side. We have spent 60 years just supplying more tarmac, and now it is time to get on with addressing the demand side: Building quality alternatives and providing clear incentives to fine-tune peoples choices.
And, just like road building, investing in quality Rapid Transit will grow the demand for more of it. It will also shift land use, incentivising agglomeration economies and greater intensification around transport nodes, as well as individual habits to suit this option more. What we feed, with infrastructure investment, grows. And vitally, inducing this sort of movement instead of driving is entirely consistent with other the demands of this century; especially our country’s new commitments to reduce our carbon emissions, and the use of our own abundant and renewably generated energy.
This project is both so expensive and potentially so valuable or so damaging that it needs a fully informed public debate about the possibilities. Gone are the days that NZTA can just keep building what its used to without real analysis of all alternatives, or that a politically expedient option sails by without serious evaluation. Because it can be transformed into a truly great asset for the city and the nation on this important route from the eye-wateringly expensive and clearly dubious idea from last century that it is now.