Last week Henderson-Massey Local Board Chair Shane Henderson (yep, great name for the role) highlighted a key issue with those criticising the proposed fuel taxes I also discussed last week.
Seeing complaining about fuel taxes, but I haven't seen a single alternative suggestion. Not a single one.
— Shane Henderson (@HendoWest) April 5, 2018
This led into a long discussion about whether road pricing is a realistic alternative. The short version of the answer is almost certainly “not for quite a long time”.
I’ll work through a few reasons why I think this is the case and why I think Auckland is probably around ten years away from being ready for a widespread road pricing scheme, noting that in the shorter term (say around 2024 when City Rail Link and Light-Rail are finished) introducing a city centre cordon scheme might be a good idea.
Anyway, let’s work through what road pricing is, what the benefits we expect from it are, whether it could be a replacement for current revenue tools like fuel taxes and what might determine the “right time” for Auckland to implement it.
What is pricing?
Fundamentally road pricing (also called congestion charging and smarter transport pricing) looks to vary the cost of travel by time or location to better account for all the costs that people impose when they travel. This sounds a bit like economic-talk, but essentially means that when people drive on a busy road at peak times they contribute to making it congested and so impose delays (and therefore costs through wasted time) on others. In a general sense demand exceeds supply and queuing, rather than price, determines who gets the available road space.
Pricing seeks to fix this by creating an incentive for some people to travel at a different time, different route or different mode by increasing the cost of travelling on the high-demand route. Because of the way traffic works, encouraging a few people to change the way they travel can have big improvements for everyone else – particularly if it gets demand back to more optimal levels where throughput can be maximised (the right side of the graph below):
Is pricing a good idea?
Talk to any transport economist and they’ll tell you road pricing is a no-brainer and they can’t believe it hasn’t yet been implemented everywhere already. Talk to any politician and they’ll tell you road pricing is political poison and they’re amazed it’s been implemented anywhere.
Ultimately pricing creates winners and losers, but if done well, should lead to better outcomes overall. There’s a reason why airline ticket prices vary by the time you travel, as otherwise you’d never be able to get on an 8am flight unless you booked months in advance. Similarly, we pay a market price for bread rather than queuing for it. Sure that means that poorer people may struggle to afford some goods and services, but (at least theoretically) we have agreed that it’s better to address income disparity directly through social welfare schemes rather than forcing bread prices down.
Putting a good pricing system in place is likely to create significant benefits. These were highlighted in ATAP, where the introduction of “smarter transport pricing” (a whole of network GPS-based scheme that doesn’t yet exist anywhere in the world) led to huge impacts on congestion and employment accessibility:
As well as huge improvements to access and congestion, pricing should result in more efficient use of existing roads and reduce our need to spend money on transport over time. After all, most roads are hugely under-utilised outside peak hours and pricing will act as an incentive for people to travel outside the peak and via more spatially efficient modes like public transport.
However, pricing systems have only been implemented in a few cities around the world and usually in a fairly crude way. The London congestion charging scheme, for example, is a flat-rate price to enter one particular part of the city during the day. It doesn’t distinguish between individual roads, peak and off-peak times (except at night) or vary according to how long the trip is within the charging area. Central London is also obviously one of the most well served locations in the entire world with travel choices, including an incredible variety of train, tube, bus and bike infrastructure meaning that the scheme only actually affects a tiny proportion of people travelling to this part of London each day.
I think the reason why so few cities have implemented road pricing is that it’s actually really hard to get right – especially with current technology.
- It’s easy for a scheme to inadvertently result in perverse outcomes (like encouraging vehicles that previously used the motorway to now use local streets to avoid a charge)
- It’s easy for a scheme to hurt lower income households with fewer travel choices
- It’s hard, in a practical sense, to vary the price that people pay to travel by time and location (at least with current technology)
So overall pricing is probably a good idea, but we should also recognise that it’s also potentially quite easy to get wrong and the implications of getting it wrong could be pretty severe.
Is pricing a good revenue tool?
Pricing is a good way to improve the transport network’s performance and will potentially raise substantial revenue (depending on the design of the scheme). There are probably two main issues with using pricing to raise revenue :
- It is an expensive way to raise money. Around 30% of revenue raised in London’s congestion charging scheme goes to operating the scheme itself, which basically means that the public needs to pay a heck of a lot more to generate the same amount of net revenue as other, cheaper revenue tools like fuel taxes and rates.
- The revenue goals may conflict with other objectives like demand management and fairness. Under a complicated road pricing scheme you could see a scenario where you “should” charge a pretty low price off-peak or on some uncongested routes to incentivise people to change the way they travel. But that’s going to reduce revenue. Furthermore, you might make a decision to only charge a low rate for travel in areas where there are fewer travel alternatives (e.g. outer suburbs or rural areas) or in lower income areas, but once again that’s going to reduce your revenue.
Overall I think you would struggle to justify doing road pricing for revenue reasons alone, because of the issues above. Furthermore, if part of your approach to road pricing is to make travel cheap at some times and in some locations, which makes sense given that the current ‘flat rate’ system over prices people who travel on quiet streets outside of peak times, then by definition pricing can no longer be about “adding more revenue” – it starts to become a replacement funding system.
So overall, I’m a bit doubtful about whether pricing is a good revenue tool, but that doesn’t mean it’s a bad idea (because pricing is a good demand management policy).
When should Auckland implement pricing?
The Phase One report of the “Congestion Question” project, formerly known as the Smarter Transport Pricing Project, draws from a wide variety of international experience to make some key recommendations about what might be the best way to implement pricing in Auckland. Essentially it boils down to two implementation options:
- Start with something small (like a city centre cordon) and expand over time as technology improves, people get used to the idea and travel choices improve.
- Do a “big bang” approach and go straight to the kind of GPS-based whole-of-network scheme that ATAP proposed.
The report strongly recommends the first option, as a far less risky approach. Risk is an important consideration as proposed pricing schemes have failed just as often as they’ve succeeded. But when might Auckland be ready to take a first step, or a second step?
Answering this question requires us to get back to the fundamentals of how road pricing works – by using price to incentivise some people to change the way they travel. This might be a change of route, change of travel time or change of mode (from car to PT for example). Ultimately, some people will change the way they travel to avoid/reduce the cost and some people will not change and pay the cost. Looking at these two groups separately we can get an idea of what will help more people to come out as “winners”.
- For those who change, there’s clearly a cost to them in needing to travel via a different route or mode, or at a different time. Otherwise they would have been travelling this other way in the first place. A successful pricing scheme will be one that gets enough people to change at the lowest possible price. Importantly, the better their alternatives are (which could mean available public transport that’s nearly as fast as driving, or an employer who allows flexible work times so they don’t need to travel during peak hours) the lower the price needs to be. If your travel alternatives are terrible (say a bus that takes more than twice as long as driving, like currently exists for much of Auckland) then prices are going to have to be really high before people start changing the way they travel. This is really problematic because….
- …for those who don’t change (which may well be most people), they “win” if the gains they get from reduced congestion are greater than the extra cost they need to pay. The higher the price, the less likely this trade-off will come down in their favour and you will see a lot of people paying a lot more money to get around just a bit quicker.
This is a rather long-winded way of saying that pricing is most likely to be successful and beneficial once Auckland has much better public transport, walking and cycling options. In fact, the better these travel alternatives get, the lower the price that will need to be charged to get the same number of people to “change” and the more winners there will be.
So when might Auckland’s public transport, walking and cycling options be “good enough” to implement road pricing? There probably isn’t an exact answer to this, but in the places where road pricing has successfully been implemented (London, Singapore and a few Scandinavian cities) the vast majority of travel was already being made by non-car modes in the areas affected by the pricing schemes.This is a pretty different situation to Auckland where car travel is a minority only for the city centre, and the vast majority in other areas.
The major programme of public transport improvements that Auckland will implement over the next decade (City Rail Link, light-rail, Eastern Busway etc.) will radically change the availability of high quality public transport for many parts of the city. But even then it seems that it will probably only be the city centre that is getting close to having similar travel characteristics as places where pricing has been successfully implemented.
Ultimately I think it is the timing of these major public transport projects that will determine when it “makes sense” to implement road pricing in Auckland. If we do it too early, before these projects are in place, we will need to charge such a high price to achieve sufficient modal shift to reduce congestion that most people will have to pay more than the travel time they save. The loss for those that have to shift the route, time of mode they use may also be substantial – and most likely felt by lower income households in poorer outer suburban areas.
Of course, this means we need to find a way to fund these public transport projects that help get Auckland ready for pricing. And that requires revenue-raising tools that we can implement quickly, and cheaply. Which at this point means fuel taxes.