Today’s AFR (subscription only, and not yet accessible on Factiva, but on p 10 of the print edition) reports the head of major private transport companies Toll Holdings opposing the privatisation of rail infrastructure, on the grounds that “
if one of those infrastructure groups got hold of the rail network [owned by Macquarie Infrastructure Group, as noted in the report-JQ] it could go the way of Sydney airport. We could see costs jacked up
Mr Little also said that private infrastructure groups would not invest adequately.
I made some similar points in my piece on Telstra a couple of weeks ago, and in my latest piece (over the fold), but this is the first time I’ve seen a large private corporation acknowledging the poor performance of privatised infrastructure networks.
Melbourne’s Scoresby freeway project, now the Eastlink tollway, has caused plenty of grief to both sides of politics. Federal Labor’s political woes of 2004 were worsened when the Bracks government announced, repudiating previous promises, that the project would be a toll road. Now Opposition leader Robert Doyle has been forced to abandon his own pledge to remove the toll.
Although Scoresby is commonly described as a $2 billion project, Doyle was advised that buying out the toll would cost at least $4.3 billion. The Bracks government published an even higher estimate, of $7 billion. Although the basis of these calculations was, as usual, not clear because of commercial confidentiality, it appears that the $7 billion is an estimate of the present value of tolls, discounted at the government bond rate, while the $4.3 billion is an estimate of the market value of the project, taking account of the additional risk borne by private investors.
This is not the first instance of this kind. The Carr government in NSW came to office promising to buy out a number of toll road contracts, but found that it was effectively impossible to do so, and had to resort to a complex system of compensating drivers. Undeterred, it has engaged in more toll road projects.
The most recent of these, the Cross-City tunnel, has been particularly controversial, since it has actually made traffic problems worse for many drivers. Apparently to protect toll revenue, other streets were closed when the tunnel was opened.
There are two major lessons in all this. The first, the consistent finding of many studies over the past decade, is that public-private partnerships are, in nearly all cases, an inefficient and costly method of financing network infrastructure projects such as roads. On top of the higher rates of return required by private investors to take risks that are more appropriately borne by governments, these projects typically involve substantial fees paid to financial intermediaries. The difference between the construction cost of projects like Scoresby and the figures quoted when considering a buyout of the toll is due, in large measure, to excessive financing costs.
Reliance on private funding for infrastructure was originally a method of getting around Loan Council restrictions on aggregate government borrowing. Despite some public disclaimers, it seems clear that the desire to deliver projects with the spurious appearance of no additional public debt remains a central component of the appeal of such public-private partnerships. In economic reality, a toll is a tax, and alienating a stream of tax revenue is exactly the same as taking on additional debt. The public is paying a high price to allow politicians to make spurious claims about ‘zero public debt’.
A more fundamental problem is that, in most cases, tolls are a perverse method of financing new road projects. The central aim of such projects is to shift traffic away from existing congested roads and onto the new roads, which are designed to have capacity for years into the future. The effect of a toll is to divert cars from the new roads to the old, congested routes. This is exactly the opposite of economically sound road pricing policy.
The need to collect tolls plays havoc with the design of the road system. Entrances and exits must be designed, not to smooth traffic flow, but to make toll collection easy.
The only sensible system of road pricing is one in which charges are based on congestion, not on the historical accident that the government was short of money when the road was commissioned. Motorists may object to paying for something they previously got ‘free’, but the central lesson of economics is ‘there ain’t no such thing as a free lunch’. What you avoid in explicit charges, you pay for in time spent in traffic jams.
One of the few politicians willing to learn the lessons of economics is London Mayor ‘Red Ken’ Livingstone, also notable as an opponent of the Blair government’s partial privatisation of the London Underground. Drawing on the ideas of the great Chicago economist Milton Friedman, Livingstone imposed a congestion tax of five pounds on cars entering central London. The result was an immediate reduction in congestion and increase in average traffic speeds. And Livingstone remains highly popular.
The Bracks government has announced a review of Melbourne’s transport problems, and foreshadowed its openness to radical policy options. Only one option has been ruled out in advance as too radical. You guessed it — a congestion charge.
John Quiggin 
I sent my proposal to establish a London-style pricing cordon for Melbourne to this inquiry so maybe (given your comments) it will not be entertained. By the way John, Melbourne’s most intractably difficult traffic problems are in the periphery not around the CBD where there are second-best issues that limit the potential for even modest congestion pricing. Australian cities are large and low density so simple public transport and road options are limited. That said I strongly agree with you that any pricing solutions anywhere are only tangentially related to cost-recovery. The target should be congestion.
Of course private firms should build expressways but governments should manage and operate them.
That CityLink charges almost uniform tolls even when congestion is light shows that it is pricing too high and not dealing with congestion. I think Scoresby will be better since prices lower and the option for improviing public transport as a competitor not removed as with CityLink.
Surely privatising rail infrastructure is going to lead to problems of natural monopoly? It was a disaster here in NZ and led to years of under-maintenance and the government (if I recall correctly) eventually buying back the tracks.
I disagree that public-private infrastructure partnership to do with transport are typically a failure. This might be the Australian experience, but it isn’t the European experience. There are tollway roads all over countries like France built in this way, and they seem to work just fine. This is a problem of how they are run, and not whether they are public or private.
There is also huge variance in how other infrastructure systems are run — whether publically or privately owned. The French train system (SNCF), owned by the government, is amazing (cheap, fast and reliable), whereas the Australian rail network is hopeless (try taking a train from, say, Sydney to Melbourne versus Paris to Nice). The same is true of private companies, where some systems are run well (like the MTR and KCR n HK, which is a weird blend of public and private ownership — they appologize when your train is 2 minutes late), and some hopelessly (and some similarly — like Melbourne trams). Thus if the public sector can’t run a decent transport system, I don’t see why that is neccesarily worse than private-public partnerships trying to run them, and indeed investing in them.
Another advantage of bringing in private companies is that it gets you around vested interest groups, which the Australian and State governments seem only too willing to pander to. Rather obvious infrastructure to construct for Australia would be fast trains (versus faster than rather slow trains) between Newcastle->Sydney->Wollongong and Melbourne->Geelong. You could also link Melbourne and Sydney, which would save people flying (travelling time would be about 3.5 hours — but you could go from the centre of the city->the centre of the city). I don’t see the government ever doing these, due to political pressure from various groups (like the airlines), and it might not be completely viable to do them privately either. Alternatively it, would be completely beneficial infrastructure to have (helping to solve population pressures in Sydney, for instance). If the only way to get is via a PPI project, that is far better than no such system. Similar examples might include things like Sydney Airport , where part of the problem is at least partially due to political ineptitude.
Conrad, At least w.r.t. roads I don’t think your claims are accurate. A firm tendering to operate a tollway has no incentives to price efficiently at congestion toll levels. You mention the good European experiences on rail — there have also been bad experiences in the UK, US and Canada tollways often associated with Transurban. Standard complaint — charges too high and not congestion-targeted. I can think of interest-group arguments against public ownership for more labour-intensive technologies (maybe rail a bit) but think this is become less of a difficulty since capital and management skill are becoming the main ingredients.
Harry, I don’t disagree that there have been good and bad experiences in different places from both public and private ownership. Paris has an exceptional public transport network thanks to a smart governement (and hence less need for toll rolls and the like), whereas London is aweful (and hence the roads are crowded, and road tolls etc. are needed as a result), due to a stupid goverment.
I also don’t believe the argument about charges is valid, since people complain no matter what — Public transport in Hong Kong is dirt cheap (probably about 50% of the Australian price), but if they try and put the price up 5c, there are people on the streets complaining.
My main complaint is that obviously beneficial infrastructure (whether they be roads, trains, or whatever), just isn’t commonly built using public funds anymore (Melbourne city loop might be an exception here) unless there is some political benefit (but it is easy to find useless projects that are). If private companies want to stick such infrastructure in and make money doing so with government help, then I’m happy for them to do it — even if it could be done using only public funds with a cheaper result and even if more appropriate targetting of things like congestion could be done. Often a poor solution is going to be better than no solution, and if Australians are too cheap/stupid/ignorant/corrupt that they arn’t willing to invest in the best solution then the second best will have to do.
The obvious example would be to stick in a fast train service between Newcastle and Sydney (or even, say, an Autobahn with charges on each end, if you like cars that much). With such a service you could travel between the two cities in 25 minutes, which is less than, say, Epping to Sydney CBD, via a normal train. It seems reasonable to suspect that if there was such a service large numbers of people would live in Newcastle and commute everyday, reducing traffic etc. in Sydney (this is what happens now in Paris — many people live in the smaller surrounding cities), and reviving Newcastle. DIgging tunnels around the place might then not be neccesary. If a private company wants to construct that and make a profit from it, this is better than no service, even if it is poorly run.
conrad wrote : The obvious example would be to stick in a fast train service between Newcastle and Sydney … It seems reasonable to suspect that if there was such a service large numbers of people would live in Newcastle and commute everyday, reducing traffic etc.
Wouldn’t it be simpler to begin planning our cities so that people live closer to where they work? Of course public transport is preferable to the private motor vehicle, but even public transport requires energy and there isn’t that much of it left in the world these days.
Rail privatisation seems to have worked well in Japan. As I understand it they privatised all their infastructure in the 1980s and they now have the best railways in the world.
Private companies are like private people in that they can have opinions that are flawed. So if Toll has some flawed ideas nobody should be shocked.
Terje Petersen wrote: Private companies are like private people in that they can have opinions that are flawed. So if Toll has some flawed ideas nobody should be shocked.
I think we all owe a debt of gratitude to Terje’s composure and clear thinking, without which the hysteria caused by Toll Holdings’ muddleheadedness over the question of privatisation might have gotten completely out of hand.
My arguments for a London style pricing scheme adapted for Melbourne coupled with congestion pricing of main arterials, reform of existing pricing arrangements and urban/public transport reforms reforms for the city’s periphery is item 3 at the VCEC website:
http://www.vcec.vic.gov.au/CA256EAF001C7B21/0/E6F4A4748F936D23CA2570840004A41A?OpenDocument
J.S. I would have thought that the problem was that we are trying to retrofit new transport sytems into cities that were originally designed 100 or so years ago! As to the most appropriate way to fund these systems would be, perhaps, an IPP with a properly costed charge that takes account of time of day and distance travelled whether by bus or car, or train if it can be fitted in.