Should rail commuters pay a congestion toll?

Melbourne’s peak train services are overcrowded and have been for quite a few years. Given the high costs that peak period commuters impose on the rail system, wouldn’t it be more efficient and more equitable if they paid more for their tickets?

After all, the capacity of the system is determined by peak demand – all those trains and the associated infrastructure and personnel required to handle the peaks are under-utilised or sit idle for the rest of the day and on weekends.

Too crowded to get a seat?

As would be the case with congestion charging on roads, a charge on peak hour train travellers should reduce over-crowding (congestion) by suppressing travel, moving lower value trips to off-peak periods and encouraging shifts to other modes. Passengers who continued travelling in the peak would make a larger contribution towards what it actually costs to get them to work.

I’m prompted to think about this issue by a proposal to levy a $0.50 per trip surcharge on customers of Washington D.C.’s Metro system who use or pass through the network’s busiest stations during the busiest period of the peak. If approved, the congestion toll would apply from next month.

Before dismissing the idea out of hand, consider the idea of reducing ticket prices for off-peak travel. That would encourage greater patronage on underutilised off-peak train services i.e. it would increase revenue for virtually no additional cost. I think most people would agree that is sensible.

But it is essentially the same idea – charging off-peak users less (than peak users) is the same principle as charging peak period travellers more (than off-peak users)! In fact peak fares in Melbourne are already higher than some fares on weekends, at night and before 7 a.m. But midweek daytime passengers pay the same as peak period travellers even though they cost less to move.

How might passengers in Melbourne react to a surcharge on CBD travel at the busiest time of the peak?

One view is that most peak passengers are captive to public transport and hence wouldn’t be able to avoid the surcharge. They travel long distances to the CBD from the suburbs by train because other modes, most especially cars, are simply not competitive for these sorts of trips. Most are travelling to and from work – hardly a discretionary trip.

Thus if this interpretation is correct, a surcharge would have only a small impact on overcrowding but a positive impact on revenues. It suggests there is little justification for peak period train travellers to be subsidised. They should pay the real costs of their travel or, alternatively, CBD businesses should pay the cost of delivering the multitude of workers they rely on to their door.

A contrary view is that many CBD workers, particularly those that work for large organisations, would be able to shift their time of travel so as to escape the surcharge. In addition, some of those who live in the inner city or inner suburbs could elect to walk or cycle instead. Further, given that around half of motorised work trips to the City of Melbourne are made by car, there may be some scope for passengers to shift to driving. These adaptive behaviours mean there would be less congestion at the busiest time.

Another take on this issue is that if peak rail commuters are really not paying their way then this is reinforcing the primacy of the CBD at a time when the Government is arguing, in Melbourne @ 5 Million, that a multi-centred city based on suburban rail lines would provide a more efficient and equitable outcome than the central city focus of Melbourne 2030. Even the Premier has an opinion on this issue.

The key risk with a peak surcharge is that it could lead to travellers substituting cars for trains. This would add to traffic congestion and be less fuel and emissions friendly than a crowded train. However given the sheer scale of the increase in demand for rail services to the CBD over the last five to six years, it might be that a 50c (say) increase in the peak fare would not make driving significantly more attractive.

A surcharge would however be politically difficult. It might therefore be a shrewd move in an election year for the Government to announce that fares either side of the peak-of-the-peak will be reduced to the current concession level. For example, the current zone one 2 hour full fare of $3.70 could continue in the tolled period (say 8am-9am and 5pm-6pm) and the concession fare of $2.30 in all other periods. There are also innovative ways in which high school students might be given an incentive to travel outside the peak-of-the-peak.

There would very likely be a hit to revenue but the pay-off would be in less crowded trains and greater off-peak patronage. Future fare increases should increase the differential between the two periods.

I’ll concede I don’t have access to all the technical information, but it’s worth thinking about. Charging motorists for road space, particularly in peak periods, warrants thinking about even more.


8 Comments on “Should rail commuters pay a congestion toll?”

  1. TomD says:

    Interesting overview of a variety of considerations, but what is wrong with the rail system that it can’t meet peak demand in the first place? Do improved practices or infrastructure in relation to its operations need to be another (or even an alternative) set of issues to address?

    • Alan Davies says:

      Infrastructure is the major issue because of years of under-investment in public transport. Also, Government did not anticipate size of increase in demand. Government is buying more rolling stock – some new trains came on line this month and consequently a new timetable came into effect on some lines.

  2. Michael says:

    Interesting post. I gave up public transport in favour of cycling partly because of the cost, but mainly because it’s overcrowded, unreliable and not particularly convenient. The main reason I would object to this idea is that I believe cars trips are already massively subsidised due to all the negative externalities they create. If those were dealt with first it might be fairer. Perhaps a land tax or council rates could have a component of public transport funding built into it. Houses or business located near public transport could help fund it since they are taking up space around it. If they use it them they benefit from the convenience, if they don’t then they should be encouraged to move away from it and free up the space for others who would use it.

  3. Janet Bolitho says:

    The massive increase in the uptake of public transport is a very recent phenomenon. The historical legacy of car oriented investment and infrastructure has a much longer history. I reckon that we have some considerable way to go in supporting pt before penalising commuters through congestion charging. Congestion pricing for cars accessing the CBD first!

    • Alan Davies says:

      Agree that congestion pricing for cars ought to be the first priority (see my last para). Note however that I’m not proposing to penalise rail commuters. My proposal (3rd last para) is to keep peak fares as they are but reduce off-peak fares.

  4. Aenveigh says:

    I assume that your definition of “offpeak” includes contra-flow passenger movements – this also acheives the goal of maximising available capacity, as at present many trains head outbound from the city with very light loadings, creating an effective trip (in and return) loading of only 50%; not so good (eg airlines achieve say 85%).

    As you’ve highlighted that suburbs house a majority of a city’s (or at least Melbourne’s) job, the question is how t o ensure PT effectively captures those trips that are suitable for this movment, and why it is not (to any significant extent) currently.

    • Alan Davies says:

      The problem with counter-flow trips is (1) not many workers live in the inner city (2) those that do tend to work within the inner city rather than the suburbs (see here) and (3) jobs in the suburbs tend to be dispersed rather than concentrated in centres served by rail.

      To really increase public transport’s share of suburban job travel significantly, what we really need is a dense grid of radial and circumferential high frequency, well coordinated public transport services. Probably buses primarily at, say, 2 km centres (or closer in inner areas). But to work on a scale that justifies the effort it would be essential that it be supported by constraints on car travel e.g. road pricing (I can feel a longer post coming on here!).

  5. Aenveigh says:

    Actually have no problem with peak charging (as a surcharge) as these are the most expensive customers to service. Express paths, peak infrastructure left idle the other 20 hours of the day, etc.
    Allows you to deepen the offpeak discount, using price to encourage travel pattern changes (although the price elasticity has limits, e.g 35% of Perth residents recently surveyed indicated they wouldn’t drink recycled water ever, even if it was provided at negative cost).
    Would only ever be politically acceptable if you were also doing the same thing to other modes subject to congestion, ie cars; but there’s precedents for that.
    One final point; the Washington model is good in that it is charging the peak central stations only; so if you’re travelling from the far out to the middle, you don’t pay, even if you’re travelling in peak hour. This is useful, as commuting the shorter distance frees up the seat for another inbound trip, ie greater seat utilization efficiency. (2 fares rather than 1 acquired for the same seat).


Leave a Reply

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

WordPress.com Logo

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

Facebook photo

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

Connecting to %s