Wednesday, May 30, 2007

High petrol prices & worsening congestion increase use of public transport

Critics of public transport pricing claim in The Age that public transport fares in Melbourne are ‘too high’ compared to other States – the State-by-State fare comparisons are here.

Indeed this claim at might be true in a comparative sense and from the viewpoint of efficiently pricing such services at social marginal cost - the latter suggests quite low prices. But despite these apparently high prices there has been a surge in Melbourne’s use of train services – about an 18% growth over the past two years.

This is interesting and reflects higher traffic congestion and petrol costs.

These factors force motorists to seek alternative modes of transport. Transport economists from other States have suggested to me that the same broad conclusions hold elsewhere. Indeed in Melbourne there is now heavy peak hour congestion in use of train services as commuters switch from driving their car to taking the train.

The conventional wisdom is that cross price elasticities* of demand for public transport are low. For example the text by K.J. Button, Transport Economics, 1993 suggests very low effects of increased petrol price on car use. This, in turn, suggests cross price elasticities of train demand in response to petrol price changes are also low since not many passengers will divert to public transport when the cost of using their car rises. The direct effects of increased congestion costs on car usage are also often seen as low so, it is usually argued, road charges to internalise congestion costs need to be high.

I’ve taught these types of ‘stylised facts’ in courses on public policy and transport economics for 20 years. But the recent Melbourne experience makes me rethink. Maybe these conclusions change when petrol costs are so high that they impact significantly on household budgets and when motorists have to drive in heavily congested conditions.

I have long advocated a case for congestion pricing in Melbourne. But this case improves when petrol prices are also high and people are upset by having to spend long periods on congested roads. Then slapping on a congestion charge – even a low one – might have very strong effects in further encouraging motorists to use public transport. Only low use road charges are then needed to tip the balance back in favour of public transport because these low charges ride on the back of other costs associated with motoring which is already increasing.


(An alternative interpretation is that with high petrol prices and long travel times the congestion problem will sort itself out. This seems fanciful to me given the steadily worsening road traffic situation in Melbourne. As some motorists leave the road because of high costs latent demands to use roads are released so public transport use can increase without congestion changing much)

Low road use prices are likely to be much more politically acceptable than high ones so it seems to me that the advent of high petrol prices improves the political case for road pricing. Of course there is an accompanying need to expand the stock of public transport infrastructure – a task the Bracks Government has in Victoria has already started on.

Thus there are indeed a few very indirect benefits from high petrol charges. Of course there are far greater economic costs since individuals driving in uncongested areas also get slugged with the higher petrol prices. This is the reason that taxes on petrol cannot generally be used as a surrogate for congestion pricing.

*A cross price elasticity of demand for train services with respect to a petrol or congestion cost measures the percentage change in train demand from a percentage change in the respective cost. It measures how sensitive rail use demands are to these cost changes.


Update: As Sinclair Davison (and colleague Rob Waschik) point out the post above neglects to mention the substantial ($800) parking charges that the Bracks Government has introduced in Melbourne. These compound effects promoting train travel.

4 comments:

Anonymous said...

Harry - there is also an $800 pa parking tax in the CBD. That with petrol prices may account for the increased public transport use. )In effect that is your congestion change. So your argument appears to hold.)

Anonymous said...

Harry.
What congestion are you talking about? Seriously we live in the same city it seems and getting around the place is a breeze compared to other places.

There's no congestion where will live in a relative sense.

And yes the parking tax is a congestion tax. Notice though how the tax was used as a revenue earner seeing there was no mitigation.

Clinton Barnes said...

A lot of stick ideas there Harry Clarke, not much carrot. Is public transport a 'market failure', should the revenue from these sticks go into making public transport something we want to use?

hc said...

cb, Hypothecation is a non-no generally in economics but I think it makes sense here. Of course we should seek lower fares, better serviuces and more community reliance on public transport. It compensates those who lose out most from such things as road pricing.