Friday 11 November 2022

Congestion pricing doesn't require public transport to be an alternative for most trips - the Mayor of Auckland makes a common mistake

Does congestion pricing need all motorists to have a "reasonable alternative", by mode of transport?


The alternatives to congestion pricing include:

  • Modal choices (public transport and cycling/walking for shorter trips)
  • Time of day choice (driving outside charging periods)
  • Route choice (avoiding driving on priced roads if possible)
  • Trip frequency choice (driving less frequency if a trip is regular)

The single biggest reason cars have become ubiquitous in cities is because they enable trips to be undertaken between origins and destinations that are not, and are likely to never be efficient to take by other modes. For trips to city centres, public transport can be an option for many, at peak times and indeed for trips between locations on public transport corridors, they obviously can be undertaken instead of driving, but for many cities, especially lower density cities like Auckland, characterised by houses on blocks of land in outer suburbs. It is not ever going to be efficient to have bus services that are always walkable to all of those suburban outlying areas.

So those that claim public transport alternatives need to exist for everyone are mistaken or actually just opposed to congestion pricing.  I hope the Mayor of Auckland is simply mistaken.


I've written before about Auckland, New Zealand's, experience in investigating congestion pricing.  A major report was released just before the pandemic (see here) and I wrote about it here.

In short it was recommended that Auckland implement congestion pricing, starting with a small inner city cordon (effectively as a pilot, but consistent with Auckland Council's objectives to reduce car use in the inner city and provide more space for pedestrians, cyclists and buses) followed by corridor based charging, in some ways resembling how Singapore's excellent ERP system was rolled out. 

The pandemic stopped further progress, but in the meantime the Government of the day was re-elected, with a majority for the ruling Labour Party, and it followed up with a Parliamentary Inquiry into congestion pricing in Auckland. which recommended that congestion pricing proceed.

It's worth noting that congestion pricing is not controversial at the level of national politics, although there is debate about how to use the money, and the priority of objectives. The Green Party supports it, so do the Opposition centre-right National and free-market liberal ACT parties.  However, it would be fair to say that progress in advancing congestion pricing in Auckland has been glacial over the past year.

To enable congestion pricing, the Government needs to introduce legislation to permit it to be introduced and this has not yet occurred.  However, in the meantime there were local elections in New Zealand. Auckland's Mayor had retired, and he has been replaced by the centre-right candidate Wayne Brown, who has reportedly now said that congestion charging is a "distraction" and 

"Congestion charging could only make sense once every Aucklander has the option of catching a bus or a train that they know will show up on time, every time – and we are two years away from that, at the very least"

There are two big mistakes with this:

  1. If it were decided to introduce congestion pricing, it would take at least two-three years before it could be ready in any case.  Indeed, the argument has been made that the right time to introduce it is when Auckland's underground city rail loop is completed in 2025 or so.
  2. There is no reason for EVERY person to have a public transport option to substitute every possible car trip, indeed it is neither possible nor rational nor necessary.  This is because many trips will simply change time or frequency of travel, especially more discretionary trips.
What does congestion pricing do?

Fundamentally, congestion pricing is applying a time and location sensitive price to road use directly, with the intention that it shall alter behaviour of marginal users of the road network (those sensitive to price relative to the value of the trip by road at that time and location), so that they do not use congested roads at specific times, reducing congestion to more economically efficient levels.

In most cases congestion pricing has not been mostly about people shifting modes of travel, although modal shift is very important, it is a common mistake to think that if, say, 5000 car trips are priced off the road at 0800 that the occupants of those 5000 cars (which will be more than one per car on average) all catch a train or bus.

In Stockholm congestion pricing reduced trips over the cordon by 25% (Source: Centre for Transport Studies, Stockholm, CTS Working Paper 2014:7), but of that 25% the percentage breakdown of behaviour change varied:
  • around 10% points were commuters that shifted mode (so only around 40% of car occupants went to other modes)
  • around 6% points were discretionary trips (likely social, recreational, retail) that changed either the destination of travel at charging times or travelled less frequently.  So they still drove, just fewer times per week (consolidating trips) or to another destination not subject to pricing.
  • around 5% of trips were commercial trips (deliveries, tradespeople, taxis) that disappeared, again likely due to consolidation of trips (deliveries being managed more efficiently, tradespeople booking work for a single trip across the cordon rather than multiple ones during a day)
  • <2% changed route (using the bypass motorway to avoid the charged cordon).
  • <1% changed time of travel, noting the Stockholm scheme operates all day, but has lower prices between the peaks. 
The key conclusion being that a great deal of behaviour change is having fewer discretionary trips and more efficient management of trips.

In London most shifted mode it is believed, to buses, but it is difficult to be sure because there was a significant uplift in the level of bus service, which is believed to have resulted in more trips in any case.

For Auckland it is proposed to only have charging during the peaks, so the potential for changing time of driving is significant, much more like Singapore, which only applies charges at times when demand slows traffic below stated performance levels (very few routes have charges outside peaks, especially post-Covid).

What would happen in Auckland?

If the first stage of congestion pricing were introduced, it is likely a significant proportion of motorists would shift mode of travel, assuming bus service reliability were back to pre-pandemic levels, and the City Rail Link is opened.  However, the majority would still drive. Of those that wouldn't, many would drive off-peak instead, some discretionary trips might go elsewhere at the peak, and some would not travel at all.  

It's worth noting that the highest peak charge was assumed to be NZ$3.50 (US$2.10). Over a work week this would be NZ$17.50, this is less than the normal public transport fares (there is currently a 50% discount applied by the Government to address the cost of living temporarily) and of course for many they also pay for parking.  

The inner city cordon is estimated to reduce car trips across Auckland by only 0.4%, and even if the FULL corridor scheme were eventually introduced, there would be a 1.3% reduction in trips.  This is enough to make a significant impact on travel times and trip reliability for the vehicles that ARE paying, plus of course buses.

It's also worth remembering that freight/logistics trips mostly cannot shift modes, and in many cases can't shift time of travel, but that isn't a reason not to price those vehicles using a scarce resource (road space) when they will benefit from it operating more efficiently.

Bus services can be improved significantly because of the effects of pricing

This is the almost instant "win" of congestion pricing, ignoring any money that might be raised and diverted into public transport. Reducing traffic congestion improves travel times for buses, improves trip reliability for buses, and enables more services to be operated with the same number of buses and drivers. FHWA (US) noted that in London:

 “Excess wait time” at bus stops fell by 24 percent across Greater London during the first full year of charging, with a 30-percent decrease within the zone itself.

Stockholm also saw such an improvement in trip times, that bus services that had been enhanced were cut back somewhat because too many services were operating with too few passengers. 

Congestion pricing CAN be designed to make life better for those driving

Many who support congestion pricing do so because they want fewer car trips, as an objective, but it is worth remembering that the goal of the original Congestion Question study was to improve network performance. Less congestion benefits those who pay to use the roads at peak time, because they have more reliable travel times, shorter travel times and also save on vehicle fuel/energy costs. Moreover, the use of net revenues can be applied to reducing or abolishing other motoring taxes, such as the NZ$0.125/l Auckland regional fuel tax (including GST), which apply to ALL road users, regardless of where and when they drive.  

Mayor Wayne Brown expressed concern that transport policy in Auckland under the previous Mayor was increasingly antagonistic towards drivers, but he could embrace road pricing as enabling better conditions for drivers and public transport users, and as a tool to reduce the burden on motorists outside peak times. 

What SHOULD happen?

The New Zealand Government should announce it is going to introduce legislation enabling congestion pricing under certain conditions. This won't force the Mayor of Auckland and Auckland Council to proceed with pricing, but will raise the question more explicitly, and it will enable Wellington (and other cities) to progress it, if they so wish.  

Given 2023 is election year in New Zealand (and polls indicate there is a reasonable likelihood the governing Labour Party would be voted out of office), it might be too much to expect this level of courage, but this IS a government that has set a target of reducing total kilometres driven by light vehicles, across the country, by 30% to meet climate change targets. To implement that requires measures much more intrusive than a congestion charge for central Auckland.

Congestion pricing in Auckland should not be delayed because the Mayor doesn't fully understand the impacts it will have. Government should enable it to happen, and progress detailed design for the preferred scheme options, with the intention of being able to implement it by 2026 at the latest (and if the Government changes after the election, it might help replace the Auckland regional fuel tax, which was implemented by the current government. 

A focus on making Auckland public transport more reliable and frequent is fine, but road pricing can help with that, and it would be wrong to lose momentum on Auckland road pricing because of concerns that can be easily addressed in the detailed design phase.