Can we ever get road user charging to work?


I’m an economist – I think I understand markets and am tempted to use them when they can do a good job.  Road user charging would do a good job of keeping people off roads at peak times and, used properly, could kill traffic congestion.

I’m reluctant to endorse road-building sprees before we even try to manage the use of the infrastructure we have. Taking, say, the least useful 10 per cent of traffic off roads could kill congestion for a while and mean we can wait another few years before making huge billion dollar infrastructure investments. That has big value to society.

(A small drop in traffic can make a big difference: While slow but steady traffic can be efficient, there is a tipping point where congestion is too high, the system overloads and it fails. At that point the effect of another road user joining the road is negative sum. We really want to be below that point. Also, a surprising share of trips are ones that could be moved around – less than a fifth of trips are commutes.)

However. Road user charging has serious problems. The big one is fairness.

Charging people to use existing roads is seen as unfair – and what’s worse, it probably is. (Perhaps not the most unfair policy in all of the Australian history but enough to pose an insurmountable political challenge to implementation.)

Yes: we charge for food, we charge for electricity, we charge for water, and nobody would deny those are more fundamental. But roads are different in a key way.

The big factor is that a group of people who are below and around the average wage consume a lot of road time, and they do so in ways they can’t avoid. Remember, most jobs are in the suburbs, which are not well served by public transport.

(This is not to deny Joe Hockey’s ‘controversial’ point that the very poorest people own cars less and drive less. He is right. I’m kind of displeased with this ABC fact-check that rates the comment misleading. It’s plainly fricking true, and he was pointing it out in the service of raising the petrol excise, which is good policy. n.b. All this is not to deny that Joe Hockey was a bad Treasurer. Also, he blocked me on Twitter for making a joke about the aforementioned debacle. The joke, shown below, is totally innocuous.)

Alack, I digress. The key point is that less well off people live further out and consume a lot of road time, so road user charging is seen as pure sadism. I wrote about it recently at Crikey and a commenter asked why I wanted to “penalise” people. I read the comments on these sort of pieces a lot and I very often see charging for roads described as a penalty or punishment.

If your policy solution is perceived as cruel, you have a big problem. I’ve written previously about a PR strategy for congestion charging (and gee it’s a good one) but I’m still not seeing much progress for the idea. Maybe, just maybe, we need more.


My Crikey piece tried to sort through the ways in which a market mechanism works to sort out why the market is seen as unfair.

A market mechanism rations demand, yes – and encourages supply too. It also incentivises suppliers to be efficient; tempts competitors who may do a better job to enter the market; encourages people to consume things that are cheaper; and also, reveals who wants or needs something most.

Well, it does that last one in theory. In reality, the buyer is not always the person who wants the thing most in any sort of objective sense. Budget constraints are the deciding factor: Rich people can afford lots of things while poorer people have to make a lot of tough decisions.

Of course, it is not just roads. Many consumer goods go to rich people even though poorer people would get more subjective value from them. This, we implicitly say in a capitalist world, is fine because the reason the system works at all is by rewarding the people doing the highest value work with a life of being able to afford more things and facing fewer tough trade-offs. (this implicit choice is an issue for another time.)

But work is not something we consume like other goods. Getting to work is not a reward for work, it is an input. If you make getting to work more expensive, you risk putting some people off going altogether.

If there is literally no other option for how to get to work, and no way to bargain with your boss for another start time, and you don’t think you can find another job, and all the more fuel efficient cars cost more than your current car would fetch second hand, then the effect of a higher road use charge is 100% income effect / 0% substitution effect and you can see why it might seem punitive.

Screen Shot 2015-07-27 at 1.52.37 pm

When household budgets are tight due to child care and high effective marginal tax rates, a road user charge might actually prevent a person from commuting. If we dissuade someone from working now, it can not hurt them only now, but damage their lifetime earnings, and the lives of the generations that follow them.

Even if most people are not at these extreme cases, the trade-offs that a road pricing mechanism forces are difficult ones over things that already seem hard.

So you can make a good case for a road rationing mechanism that tries to ration road use without putting a cash price on roads. I have previously proposed a system that might do that.

A recent article in The New York Times describes a food donation network that used a kind of pricing mechanism to solve an allocation problem. It created a big range of efficiencies via a points system — local food banks used points to bid for various items from each other.

A similar points system could overcome the fairness problem in pricing roads. People could be allocated points based on, for example, how far from the CBD they live, or their lack of access to public transport. Unwanted points could be sold off.

This system would totally eliminate congestion if few enough points were given out. It would also create incentives to use other kinds of transport. And it would do so fairly.

What it doesn’t do is reward the owners or operators of roads, or provide a funding stream for roads. Public provision and funding would still be necessary.

But perhaps waiting for the perfect market mechanism that can deliver all the potential upsides is unrealistic. Can road pricing advocates choose not to let the perfect be the enemy of the good?

The more I think about this, the more I see the devil is in the detail. Who would get the points?

This goes to another of the points I made in the Crikey piece. Market allocations of goods may not be fair, per se, but they are a sort of Schelling point – a way of allocating scarce resources that society generally tends to land on in lieu of other ideas.

I like the idea of giving them to households with worse public transport options or lower access to jobs. “I have no other options” is one of the big reasons road user charging is seen as unfair. Having no other transport options is also sometimes a proxy for socio-economic deprivation, although an imperfect one.

But an allocation based on geography would be tough for any government to operate without causing a fuss and furore that would see them voted out.


While society doesn’t agree to market distribution of road space, yet, that doesn’t mean it agrees to some other rationing plan either. We have a few non-market rationing models in place in other parts of the economy: Income testing and asset testing are both known ways of allocating scarce, government-provided goods. Neither is an obvious fit for allocating road space.

So a geography-based allocation model would be wholly new. And in our geography-based political system, very much open for abuse. New allocations of road use credits would not be safe from the electoral cycle, although whether congestion would rise to previous levels is not clear.

We should also consider the unintended consequences of such a system. Would it reward sprawl if you gave road access credits to people with low public transport access? And punish people who’ve bothered to make the trade-offs to live in smaller spaces closer in?

The whole thing starts to look very difficult indeed. Is there another way to make a non-market rationing system work? Please leave your ideas below!

Published by


Thomas the Think Engine is the blog of a trained economist. It comes to you from Melbourne Australia.

21 thoughts on “Can we ever get road user charging to work?”

    1. That’d be a good option to preserve the substitution effect with creating so much of an income effect, (albeit hard work distinguishing commutes from other types of trip…).


      1. People who travel for the purposes of work (rather than travel to get to work) already deduct those travel costs from their taxes. Shouldn’t be a bridge too far if people estimated the costs of their mostly daily travel into work — especially on public transport with its set figures each day.

        A larger practical problem is that income tax is collected federally and the congestion tax would be collected at the state (or possibly local) level.

        Liked by 1 person

  1. What’s wrong is our loss of control over the / our democratic process – you allude to this in the article. And where you want to debate it your voice is blocked.

    Roads are like banks, power, post and rail. Something else to be sold off for short term Party gain.

    Then you have to create another peoples bank. Or put up taxpayers money for power inputs to fight the exorbitant uncontrollable profits of power suppliers and build more dams, provide battery sinks and new gas generators to regain some control over what politicians have sold off and I suspect we wanted kept public. Or pay exorbitant CEO wages to have our letters delivered when the people sold off something on the basis it would be better run.

    Even co-operatives and other people (not Government) owned businesses that we were talked into corporatising; AMP and NRMA being two that come to mind are examples of the same problem of loss-of-control and out-of-control.

    It doesn’t answer the question of how to decongest roads. The answer to that lies in the majority taking better control of out democracy then we will have a chance to make fairer equatable societal decisions.

    Liked by 1 person

  2. Thomas, Thomas, Thomas. For one who is normally so rational, and careful with the facts, your assertion that:

    “This, we implicitly say in a capitalist world, is fine because the reason the system works at all is by rewarding the people doing the highest value work with a life of being able to afford more things and facing fewer tough trade-offs”

    is two fifths bollocks, though the general point is well made. What the western suburbs tradie or sales rep understands only too well is that (a) there is no acceptable substitute for the best local main route to work (as you point out), (b) supply of new alternative roads is so sticky the whole idea of elasticity of supply is one the tradie / sales rep and his/her partner driving to childcare would die laughing about as you explained it; and (c) and my real beef, is that in a merit based society where hard work and ability ought count for a great deal the facts of the matter are that the tradie / sales rep cannot effectively compete with the children of their northern, eastern and southern suburbs neighbours who have access to the accumulated wealth of their parents, whose wealth was accumulated largely through tax concessions and is effectively denied to the tradie / sales rep in perpetuity. Now, would you like to do something about that? If so please get in touch.

    Liked by 1 person

    1. Ha! The big question. Does our meritocracy reward merit? Or something else?

      There’s lots of hard workers earning a pittance and some lazy screwballs making millions from doing next to nothing.

      Seems to me the rewards accrue when you work hard to own things and/or exploit markets where competition is thin, rather than when you hard work in a job.

      And what people will pay for vs what people truly value is a whole other problem.


  3. There have to be more targeted ways to improve the lot of poor people than encouraging everyone to drive everywhere at all times of day. Moving income tax brackets for example.

    Congestion charging could fund a massive increase in bus services along the now free-flowing streets (as was achieved in London). This in turn can make it feasible for many more people to live without a car, thus cutting out those major costs.

    Liked by 1 person

    1. Oh, for sure. This is a fifth-rate way to help the least well off, but maybe a second-best fix for moving us toward a situation where roads are not overused so much that they stop working.

      I like your second point of hypothecating road charging revenue for building alternatives. The problem there is the time lag. While the charging starts today, the new services are still years away and at the mercy of several new state or federal governments before they happen.


  4. What has eternally annoyed me about road pricing is when I am paying my $2.50 (or higher) to sit in traffic moving 40-50km/h when the speed limit is 100km/h. Someone on the supply side is not holding up their end of the deal. So how about this for an idea: charge the toll at the percentage of the nominated speed limit; thus, if I was going at 40km/h when the speed limit was 100km/h, I’d only be paying 40% of the toll. This then also provides incentive on the supply side.

    Liked by 2 people

    1. Late to comment on this but worth noting… the underlying point here is using a completely incorrect wrong model of congestion and supply. Jason made a similar error in the text, stating that a road has a carrying capacity and slows when it exceeds it. As Michael explains below, this isn’t what happens. Congestion is largely caused by queuing at the exit of a road (particularly on a tollway) because the throughput onto the stretch of road leading to the intersection is lower than the throughput through the intersection.

      So the “supply side” in this equation is not the toll road you paid to get on, but the congested urban streets you are trying to get off. Lowering the toll would only make this worse (more vehicles entering the tollway) without any means to make it better (the streets are what they are).

      Around the CBDs of Australian cities the massive increase in inner urban housing means that congestion on freeways is only going to get a lot worse. The streetspace is going to have to shift to more efficient uses (p/t and walking) which will further lower the capacity. On a practical point regarding road charging, one worthwhile shift would be to remove tolling from entry points and put it on exit points (plus places where the road narrows – bridges etc.). That would have the benefit of shifting trips from inner-urban orientated commuting that are better suited to p/t to bypass trips which is more likely to be freight and service vehicles.

      Liked by 1 person

      1. “The streetspace is going to have to shift to more efficient uses (p/t and walking) which will further lower the capacity.”

        To be pedantic, that shift will *increase* the capacity that matters – people – while reducing the capacity of motor vehicles.

        Reducing motor vehicle capacity should be an explicit goal of our transport policy.


  5. The whole pricing poor people off the road and regressive tax aspects are obviously important, but needn’t be an essential part of how a system works and delivers benefits. The keys are (1) time-varying charges and (2) indentifying where revenue goes.

    (1) With time varying charges, people who are less willing/able to pay can still travel at an alternative time. Because many areas have bottleneck type constrained capacity, this means that travel time gains come mostly from queuing reduction and so ‘traveling earlier’ means mostly departing at the same time as now for those who can’t afford to pay. Technically you don’t raise the generalised cost of travel. Assuming people need to be priced off the road completely is a fallacy that results from thinking of travel as a static phenomenon, without accounting for time of day aspects.

    (2) A well designed road fare system should probably spread revenues over three areas. (a) reducing registration, which is a highly regressive tax. (b) improving public transport services, which tend to be progressive, and (c) going into some kind of road building fund (linked to expanding capacity). A system like this should be progressive, not regressive. By identifying this in advance, you don’t have the money black hole effect, where the concept gets thrown out due to percieved unfairness before consideration.

    Lastly the biggest issue with your proposal of geographic credits is that the transport market is intrinsically linked to the housing market. The biggest beneficiaries of the improved access road fares would provide are the outer suburbs, so house prices will likely increase there, benefiting landowners in the outer suburbs. A related factor is that housing markets will slowly change with better road pricing, and that providing geographic credits would undermine this long term change and benefit.


    1. Thanks for the comment Matt. Can you explain more what you mean in point 1? How are you generating queuing reductions and having people leave at the same time? Is there an assumption extra infra is built?
      Re: the point of not raising the general cost of travel: Would you subsidise off peak traffic to average out the rise in peak trip costs? Surely road travel is underpriced on average and the price needs to rise on average?

      I like your points about what to do with the revenue and your argument about house price rises. This proposal leaves the question of value capture untouched. If you haven’t seen it, however, the Grattan Institute put out an interesting report on Value capture the other day saying don’t bother, just have a broad land tax:


      1. “How are you generating queuing reductions and having people leave at the same time? Is there an assumption extra infra is built?”

        Probably better explained by others, e.g. here:

        No extra infrastructure built. Generating queuing reductions, but having the same arrival profile of people as now (not leaving at the same time). As many roads/areas act like a bottleneck (e.g. CBD access), with a fixed capacity of arrival rate, you shift the current queuing cost (paid in users time) to a monetary cost (an economic transfer).

        “Re: the point of not raising the general cost of travel: Would you subsidise off peak traffic to average out the rise in peak trip costs? Surely road travel is underpriced on average and the price needs to rise on average?”

        My point was that at the moment people currently pay with travel time, and schedule delays. This has a cost. Dynamic tolls convert the queuing/excess congestion cost to a toll, and so don’t raise the total *generalised* cost of travel. So no, I don’t think that the generalised price needs to rise, though of course the monetary costs in the busiest part of the peak period do.

        Subsidising off-peak travel is pretty much equivalent to peak hour pricing, I think the Dutch government is currently experimenting and researching this.


  6. This is likely to become a bigger issue as electric cars start increasing in their share of traffic, without fuel excise and the promises of lower incremental operating costs they could have even less incentives to make space on the roads.


    1. YES! The erosion of fuel excise is going to be massive in a few years. Really important to get some sort of road space x Axel weight x per km charge into the public discourse before everyone has electric cars and it becomes a mass hip pocket issue rather than a niche tesla/prius issue!


  7. Land far from public transport is cheaper than land closer to public transport, so there’s no need to compensate people for their lack of transport choice – they’re paying so much less for housing!

    If there’s an equity issue it’s around zoning and NIMBYism that unnecessarily limits the amount of people that can benefit from living closer to the centre.


  8. Good article mate. Hope you’re well.

    My problem with road user charging is that it is unfair because users don’t have real alternative options to travel. If you had adequate public transport and then introduce congestion charging it makes sense. In the absence of adequate public transport you’re penalizing road users who don’t have a choice and have to use the road and then pay for contributing to congestion.

    Regards, Shishir

    Sent from my iPhone



Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s