A sample text widget

Etiam pulvinar consectetur dolor sed malesuada. Ut convallis euismod dolor nec pretium. Nunc ut tristique massa.

Nam sodales mi vitae dolor ullamcorper et vulputate enim accumsan. Morbi orci magna, tincidunt vitae molestie nec, molestie at mi. Nulla nulla lorem, suscipit in posuere in, interdum non magna.

Is it really time to impose congestion charges?

These are some notes I prepared for the SMART Workshop  on “Infrastructure economics and policy: new tools for new problems” that was organised by Henry Ergas and held last Thursday and Friday at the University of Wollongong.  My endorsement of Ken Henry’s urging a year nor so ago for a more responsible attitude by economists towards commentary at the stage where policy was being formulated won me no friends. But I’ll stick with it. 

The question I have been assigned is specific. It does not mention other vital tax and charging reforms in the road sector such as heavy vehicle charging and sensible insurance charging for traffic accidents that are closer to becoming immediately feasible.  Nor does my assigned topic mention reforming existing charges, such as the fuel excise, although one can drag that in by arguing that the case for congestion charging could be assisted by simultaneously abolishing the excise as a type of tax neutral deal. This latter argument is not one I buy anyway as I am fond of the fuel excise as a useful revenue earner that can be collected at low cost, which is progressive in impact and which encourages substitution away from a fuel whose abundance the market still myopically exaggerates.

Generally moving towards an efficient road transport system is something that makes sense for a vast country like Australia with concentrated urban populations.  But this is a much broader task than arguments for congestion pricing.

The immediate case for congestion charging centers on the cities of Sydney and Melbourne where there are substantial deadweight economic costs associated with congestion that will double through to 2020.  These are large sprawling cities with relatively compact CBDs but with dispersed surrounding urban and commercial areas.  Significantly, there are significant congestion problems in areas distant from their CBDs and this makes the design of congestion pricing schemes more difficult.

Limited approaches to congestion pricing such as CBD cordons face a number of difficulties whether they operate with or without pricing of major ring roads and arterials.  There are problems of addressing parking in areas on the boundary of such cordons but these are overstated and are no big deal.  The key issue is that partial schemes do not immediately deal with congestion pricing and ‘rat-running’ byproduct distortions in the city periphery.  Take the specific example of Springvale Road in Melbourne which is regarded, in RACV-conducted polls, as including the worst congested parts of the city.   This road has had pressure taken off it by the tolled EastLink cross-town road but is still heavily congested most of the day.  There would be almost no effect at all of a CBD-based cordon on this outer-suburban congestion since most of the traffic here is cross-town.  The same remote-from-the-CBD types of problems occur around Melbourne and Sydney as well.

There is the need for a comprehensive congestion plan that specifically addresses problems in the city periphery. This might draw on GPS or other pricing technologies with access to the CBD being priced at one rate and travel in the periphery at a lower rate or it might involve ‘third-best’ public transport reforms as well as parking policy reforms.

It is quite clear from the fuss over the decision to price Eastlink and the decision to not price the Frankston bypass in Melbourne that our spineless, state politicians anticipate huge political costs in any move to congestion price.  These fears have driven a ‘scardy-cat’ attitude on the part of their attendant civil servants and the consultants who advise them.  Given the fixed costs and political obstacles associated with any sort of reform I am more favorably disposed to finding a comprehensive solution that prices the lot rather than a restricted smorgasbord of policies that we know in advance will not work.

That is not to say that particular auxiliary policies cannot be usefully employed.   Donald Shoup has revived interest in parking policies and suggested that in some situations such policies can substitute for congestion pricing.  However policies that promote market clearing in parking markets are desirable with or without congestion policies so I would favor these anyway.  They are a useful though imperfect alternative to cordon-based pricing and can help address congestion in the periphery as well if implemented there.  Park-and-ride policies (with free parking near train and bus stations) are expensive and illusory ways of addressing CBD-based congestion – the parking should at least be priced.   Generally encouraging local governments to price parking at market-clearing levels, discouraging free parking in shopping centers and discouraging parking-cum-residential entitlements in apartment contracts will improve congestion in the suburbs by making people more aware of the costs of parking. These measures simultaneously have attractive incentive-compatibility characteristics for revenue-starved local governments.

So what do I suggest?  Without action the congestion problems of Sydney and Melbourne will become acute over the next decade.  It is already so serious that doing nothing should not be an option.  But given manifest failures in our state political systems and in the way economists, public bureaucracies and other interest groups can be expected to behave I think several issues need to be thought through if sensible policy outcomes are to arise.

  1. We must think about the implementation and political acceptability issues of congestion pricing. There is now a huge literature on these issues and much more international experience to draw on than there was 20 years ago.  The technical issues in choosing particular pricing technologies are second-order cost-benefit issues. Far more important is to consider moves that will increase community support for congestion pricing. These include prospects for trialing schemes, organizing referenda to seek trial approval, appropriately timing reforms in the political business cycle,  offering commute alternatives, offering bribes to commuters as well as the standard issues of appropriately compensating the tolled-off. All of these issues need to be carefully examined.
  2. We should think about our role as economists as we intervene into political debates.  Sensible reforms to mining taxes and advocacy of the carbon tax proposal have been subject to nitpicking, irresponsible criticisms by economists that have been counter-productive in delivering good social outcomes.  On the plain packaging of cigarettes and gambling tax reforms more sinister commercial drivers are presenting interest group arguments as being in the social interest. There are venues for discussing policy alternatives but, as Ken Henry suggested two years ago, there are times – times when operational policies are being put in place – when economists interested in actually getting something done might well shut up.   It will be difficult to filter sense out of a congestion pricing debate if it is debunked as “another big tax” or rejected on the grounds of second-best effects of pricing on preexisting distortions in labour markets. These are important issues but there are appropriate times to raise them and appropriate times to, as I say, shut-up.
  3. We need to prepare for the silly questions that are always raised in relation to congestion pricing.  A particularly sneaky one is that public transport reform is a prerequisite for congestion pricing.  It isn’t – though we do need to upgrade our public transport infrastructure and improve the range of bus and mini-bus services.   The congestion objectives that are practical – shifting 5-10% of motorists away from peak period travel can be met by a vast range of substitutions of which modal shifts to public transport are but one element.  I think too the argument that congestion charging will harm low-income workers who need a car to get to work needs to be carefully thought through and disposed of.  Almost every efficiency reform in history has some negative income distributional effects. It’s a dud argument given that it is the total impact of the tax transfer system that matters not the impact of particular taxes.  And Joe the plumber probably earns more than a university professor anyway and has exactly the same need to sort his journeys into low valued trips that he can substitute away from and high-valued trips that he will be happy to pay a price for.

In Australia today we face a political stalemate in terms of realizing microeconomic reforms.  Nervous nellies line the halls of power and opportunists have become prominent in parts of the public academic debate.  I guess it is not a promising start but we have undertaken significant microeconomic reforms in the past and the road sector is the most important remaining area for overall microeconomic reform.


2 comments to Is it really time to impose congestion charges?

  • Simon

    How about calling it a “decongestion charge”? The idea is that motorists are charged money, but they get something in return – a decongested road. It’s similar to buying any product.

    The key fact to get across is that this is the only way to get a decongested road. Widening roads doesn’t give you that, and nor does building more of them, at least not in the medium term. It’s only by being charged money that motorists can get what they desire – a decongested road.

  • hc

    Simon, That might be a sensible framing move – accent the positive.