To reduce congestion, charge by the kilometre
The hip pocket must be where road pricing reform commences, writes Professor David Hensher.
The call for a congestion charge is getting louder and more frequent in many countries, as major metropolitan areas experience increasing levels of road congestion. This is often accompanied by a recognition that governments need to find new sources of revenue to maintain existing road networks and to invest in new transport infrastructure.
Reform of road pricing is almost certain to occur at some time in the future. But a key challenge is in selling the idea to the community of road users, as well as a whole raft of interest groups that influence the views of society and politicians.
Simply announcing a need for a congestion charge (often misleadingly called a tax) does little to progress the reform agenda. We need a carefully structured demonstration of what might be done to introduce adjustments in road user charges that both reduce the costs to motorists, while ensuring no loss of revenue to government.
My colleagues (Professors Michiel Bliemer and Corinne Mulley) and I suggest that this can be achieved by lowering vehicle registration fees together with a distance-based charging regime. Such a package can deliver financial gains to motorists with potential for revenue growth to the State Treasury of Sydney.
This reform package is predicted to result in changes to total annual kilometres of travel (especially in the peak), and flow through gains in travel time that would reduce traffic congestion. In Sydney, this would be a 4.7 per cent reduction in peak kilometres driven.
We see this as an essential first stage in gaining community support for road pricing reform. Proof of cost reductions associated with improvements in traffic congestion can then be used to continue the reform process.
This road pricing reform plan would require drivers to purchase an on-board unit (approximately $50 one-off cost), that will record the kilometres by time of day. The off-peak kilometres are not charged, but peak kilometres will be charged at the agreed rate.
This scenario implies that if a unit is not installed, all kilometres will be charged as peak kilometres. So there is an incentive to install a meter (with the expectation that all motorists will do so), just like households have had with off-peak electricity meters or with water meters when they were first introduced.
Modelling the change
To establish the financial implications of alternative combinations of a peak period distance-based charge and discounted annual registration fees, we have modelled different options for areas within Sydney.
The key inputs, for each area in Sydney and status quo (before situation) are: the mean annual kilometres; the proportion of kilometres in the peak periods (AM and PM); the average daily cost per driver (comprising fuel and tolls, distinguished by peak and off peak periods); annual registration fees; and mean direct elasticities of peak and off peak kilometres with respect to usage costs.
For the reform scenarios, we considered a distance-based charge varying from 2 cents/km to 10 cents/km in the peak. We also allowed annual registration fees to vary from 30 per cent to 75 per cent of the current annual fee.
The system calculates the current total costs and kilometres for all drivers and revenue to State Treasury, distinguishing outlays and receipts for the peak and off peak periods.
We then introduce the range of peak-period distance-based charges and discounted registration fees, and calculate the combination of these two cost outlays for motorists. We found that each option results in both a reduced mean cost to motorists and no loss in revenue to State Treasury.
We also expected to obtain different distance-based charging levels for a given discount on the registration fee, and indeed that is what happened. The range is 3 to 8 cents/km. Taking the lowest value would ensure net gains to each motorist, but would result in the loss of neutrality (or better) to Treasury revenue.
But placing different charges on motorists over the metropolitan area would raise clear concerns from many perspectives, including the political ramifications. A preferred solution is to take a system wide approach. That is, to identify a single distance-based charge, given a discounted registration fee, that achieves the required financial outcomes for drivers and State Treasury.
The selected peak period charge is 5 cents/km with a discounted registration fee of $185, slightly greater than a 50 per cent reduction.
On average, a driver saves $9 per annum and Treasury gains $32 per driver per annum. These are extremely low amounts per driver, but they translate into sizable financial gains to all drivers and to State Treasury.
David Hensher is the professor of management and founding director of the Institute of Transport and Logistics Studies at the University of Sydney Business School.