This is a Guest Post by regular commenter Patrick Reynolds.
In 2007 the Transport Minister Labour’s Annette King formalised NZ’s excise tax on petrol as a hypothecated tax for the National Land Transport Fund. This means that the tax is only to be used for funding transport. Details here. Including this statement:
“This decision ensures that funds collected from road transport users will be used to fund activities which benefit those same users, whether the activities concern roading or public transport.”
But also this, which shows the source of the change:
“Road users have been complaining for years about the lack of full hypothecation.”
Interesting then in the debate in Australia over the introduction of a Carbon Tax that a justification of leaving transport fuels out of the new tax is because existing petrol taxes are already a kind of carbon tax. I suppose in as much as a higher petrol price discourages driving that makes some sense but not if the money raised is used to extend the road network and encourage more driving. If this money was then used to build alternatives to carbon intensive transport that idea could be more justified. So in fact it’s a bit of a jumble in Australia as deputy Greens leader Christine Milne says here:
“At the moment we’ve got a mish-mash of policy … we have got an excise regime and we have no carbon price as such.”
My aim with this post is simply to suggest that what we are observing here is the transition of the philosophical justification for fuel taxes or, indeed any kind of road pricing, away from the idea of ‘user pays’ to an understanding of them as a type of Pigovian tax. One that is designed to change behaviour and fund externalities. And not one that is primarily there to extend the network. Milne again:
“We want Australians to drive less, and when they do drive, to drive more efficiently.”
This shifts the underpinning of the tax from a funding mechanism for a sector [roads] to viewing petrol as in some ways like tobacco or alcohol: A ‘bad’ to be discouraged, as described here in the Victoria Transport Policy paper from Canada.
One way of designing a Pigovian tax is to redistribute to money back to society evenly in order to make it revenue neutral. In this model everyone would receive an equal tax credit or grant from the tax take and because only those who bought fuel are paying the tax obviously this amounts to a transfer of money from users to non users, and those who use less. It therefore rewards people who are able to arrange their affairs with less fuel use. Currently our Pigovian taxes [tobacco and alcohol] go into the general fund to pay things like Kindy teachers salaries, Ministerial travel, and prisons etc.
My suggestion in the transport sector is that this understanding of the role of the fuel tax should be used to spread a higher proportion of the revenue towards investments that allow NZ to move away from dependence on fossil fuel based transport. In short this a charter to increase the proportion of the NLTF used to fund public transport projects, especially where this is powered by electricity, as well as the more energy efficient rail freight and coastal shipping. This then has the double effect of discouraging the ‘bad’ activity and helping to provide alternatives. This recognises that transport of people and goods is itself a ‘good’ and it is not that which is desired to be discouraged.
Note that this does not include a suggestion of increasing the current level of taxation just shifting the understanding of how best to use it. Hypothecation is simply a policy and can be changed.
Processing...
Agree with your post, but remember pigovean taxes need to be justified based on real, if estimated, costs. This may or may not be a large proportion of the current fuel tax take, but it needs to be robust and defensible. Lowering dependence on fossil fuels is not an externality per se that is more direct economic management/ paternalism.
I think that hypothecation is a slightly flawed concept in ay case. The implication of hypothecation is that road users actually own the assets collectively, and are not just paying for use, but are paying a type of membership levy. In this way all revenue is theirs to spend colectively on more roads. Road users have paid for the construction and maintenance of roads to a certain extent, but did they pay the government for all the land on which the roads sit? When did road users buy Queen St for example? When new roads cut through existing crown land do road users pay the crown for the land (honest question I dont know the answer)?
We havent made similar arguments with other capital investment undertaken by the government. Electricity users do not collectively own the generating assets collectively, they are run to make a proft/economic return on investment. Why should it be any different for roads?
No point doing a market-based analysis if you have no fundamental faith in the market.
Markets fail when it comes to transport, particularly road transport (though all modes have a measure of market failure). But don’t double and triple count the costs.
And development of transport does tend to develop the productivity of an economy, leading to higher economic activity overall. Again, don’t double or triple count the benefits.
It is justifiable to collect revenue from drivers (and train/bus passengers) to alleviate congestion, as the highest economic use will pay the fee and lower economic uses will find a more cost-effective way of achieving the same goal. But don’t double count. Once the congestion fee is collected, it’s gone.
Pollution is also a side effect of transport, depending on propulsion source and tech. But again, collect the offsetting fee and then it’s gone. Can’t collect it twice.
Benefits from economic development are captured in that development (and often in the land values). Again, collect that money once (eg land values through land tax, other economic activity through income or sales taxes). Don’t keep trying to collect the money many times.
Road transport is two things – subsidised; and opaque. The opacity is worse because it leads stakeholders to end up in arguments over who has paid.
The time has come where GPS devices can be installed in every registered road vehicle that can track its use of public roads, and which ones at that. Something we’ve never been able to do before – track the exact usage of road to vehicle, including time of day and route.
The other problem is intergenerational. Some stakeholders claim “I have ‘already’ paid for this road” by virtue of silly comments politicians have made in the past. And it is not clear in many cases whether the road is ‘preceding’ or ‘reacting’ to the demand it is supposed to serve.
We have an anology in social security payments, where a lot of old people claim their pension was ‘earned’ at some point (and pre-WWII politicians may have led them to this view, when a notion of contributory/insurance model was popular).
Remove the opacity first, then the subsidies. Only keep the subsidies where they address: congestion; pollution; transfers of wealth that the tax system otherwise doesn’t pick up through activity taxes eg some land value increases.
But always start from the point of view that there is a market for transport. It fails, but once the failures are corrected, the process works.
What I enjoyed most about this article: Seeing the words “Transport Minister” not followed by “Stephen Joyce”, even if it was a former transport minister. I felt all warm and fuzzy when I saw that.
So we can end up with very cost-effective projects missing out on money because that activity class has been exhausted
Quite. I favour a very technocratic approach to these matters, except where compelling democratic or other considerations intervene. Thus, a small intersection upgrade competes with a Road Of National Party Significance. You do this in the same way as health systems rank diverse treatments with diverse costs and outcomes – by ranking each intervention for its ability to address issues, and then assigning a score. Thus, you might score one thing at 35, and another at 90 (on a 100 point scale). You do all the highest until the funding runs out. If there are other compelling reasons to upgrade a project on the list (say for example a community or community of interest _really_ wants something), then you add some more to their score, and another thing is deprioritised. It seems there is little, if any, systematicity in the current arrangement, and this produces outcomes which are well below optimal.
As for shifting the balance, that’s obviously a matter of pricing externalities adequately, and making interventions where necessary – you might do this more directly by assigning environmental impacts a value in the scheme proposed above, where currently they are merely a consideration.
Nicely put George, of course, setting the parameters of the scoring system is open to influence, like the ‘assumptions’ in MoT evaluations, like how NZTA ignores likely changes in oil price. But if at least these matters are quantified they can then be visible and argued over.
Good post – but it’s not really a choice between “user pays” versus “pigouvian taxes.” The latter is intended to just correct for unpriced externalities that mean that the private marginal cost of driving is lower than the social cost.
So a TRUE user-pays system would seek to charge the full social marginal cost to users, including costs of carbon, congestion, and any other costs.
Finally question – NZ’ ETS includes transport fuels, but you’re saying the revenue from that charge goes back into the road network? Hahhahahahaha. Only NZ could come up with something that stupid.
Thanks Stu, and no the NZ ETS scheme is not what I was referring to, but rather the Aussie idea that fuel shouldn’t attract a carbon tax because it already has an excise tax, and that this is a mad one step forward one step back situation because those taxes then go to extending the network.
Are you aware of any work that looks at trying to quantity the full externalities of road system like ours? My experience is that if something is hard to quantify it tends to just remain out of the math, certainly by politicians, but also by economists.
I have to say that NZTA’s Economic Evaluation Manual actually does pretty well in this respect. For example, it includes:
1. Decongestion benefits per km by mode
2. Health benefits from walking/cyling
3. External enironmental costs of driving
4. Reduced parking demand from PT
5. Agglomeration economies (even if this is a little limited – mainly because they’re hard to quantify)
One thing that is daft is that new road capacity does not consider the costs of additional demand for parking, which is completely inconsistent with #4 above. So economists are trying and doing better – which I think is one reason why benefit cost ratios for PT projects are increasing relative to road projects. But there’s some ways to go, you’re right.