One of the big questions about Public Private Partnerships is what happens, when the deal needs to be renegotiated in some way, 10, 20 or 30 years after it was signed. This story about changes to an interchange affecting CityLink in Melbourne is of interest. The deal is being financed in part by replacing some payments due to be made by Transurban (the Citylink operator) with a smaller upfront payment. The financial arrangements are too complex to permit a clear assessment, but one point is striking.
Last year Transport Minister Peter Batchelor said the Government wanted $200 million for the upgrade. Transurban said it wanted to pay only $150 million – the eventual figure.
So, starting with a a $50 million gap between the initial positions, Transurban got the whole $50 million and the public got nothing.
Related to this is an interesting kerfuffle over the Scoresby (Mitcham-Frankston) motorway. As many will remember, the Bracks government initially promised not to impose tolls, then reneged, copping a lot of flak in the process. The Opposition leader, Doyle, has promised to renegotiate and remove the toll. This has presented the Bracks government with interesting incentives. Usually governments involved in PPPS want to stress what a good deal they have got for the public, in terms of the toll revenue that has been promised the private ownership. But now the situation has been reversed. Faced with Doyle’s promise, the government and its agencies commissioned a PwC report[1] which said that scrapping the tolls would cost $7 billion, with a cost of $4.5 billion for buying out the project. By contrast, construction cost is about $2.5 billion plus “other financial costs” of $1.3 billion.
It appears from these reports that the value of the tolls that have been alienated is nearly three times the cost of building the road[2]. Of course, the government’s report has been produced under pressure to make the repurchase option look as unfavorable as possible. But then, the vast majority of published analyses of PPP projects suffer from the opposite bias.
What’s even more striking about this is that, in PPP circles, the Mitcham-Frankston project is being touted as a huge success, evidence that we are finally getting these things right. Something does not add up here.
fn1. The conclusions have been released, but the interesting bits like the traffic projections are, as usual, commercial-in-confidence. At least, when I asked for them, that’s what I was told.
fn2. Using the $4.5 billion buyback cost and accepting that some of the “other financial costs” would be incurred under standard procurement would give a less extreme result. But the $7 billion number is the one the government has been touting.
John Quiggin 
Of course another issue raised by PPPs etc. for transport infrastructure projects is what happens if future governments and parliaments want to seriously change the modal mix in ways which reduce motor traffic volumes and hence toll income for the private operator? To what extent is the contracting government legally restricted from pursuing such a policy? If traffic volumes start to fall due to secular trends independent of government initiative (e.g. rising petrol prices, individual commuters taking their environmental responsibilities seriously, etc.) is the contracting government obliged to implement policies to counter these trends? And what happens if traffic volumes and toll revenue are threatened as a result of decisions by: (a) Local Governments; (b) other State governments; (c) the Federal Government; (d) international treaties to which the Federal Government is a party?
$7 billion appears to be the undiscounted sum of the payments that the government would have to pay the project ownwers over the life of the road. $4.5 billion appears to be the discounted sum of those payments, plus buy other back costs, if the government bought out the project after it was built but before any cars travelled on it.
That’s a plausible interpretation, Milton, but if so, the Opposition ought to have torn the government to shreds for presenting figures in such a misleading way. And PriceWaterhouseCoopers ought to be ashamed of themselves for letting it stand.
Paul, my understanding is that CityLink has guarantees against policy changes, but that in the Mitcham-Frankston case, the private party is bearing the risk, presumably in return for higher payments
Pardon my cynicism but isn’t this just another instance of privatising the profit and nationalising the risk when it comes to PPP, BOOT etc and roads.
I recall the Productivity Commission had something to say about roads being the least suitable kind of infrastructure for PPP/BOOT financing.
JQ, can we generalise this into a discussion of why we have to pay tolls where the last bit of road was built instead of where traffic congestion is worst — and what PPP/BOOT schemes have done to impair sensible road system management for decades into the future?
For background on transport planning, I’m mightily impressed by this site:
http://www.vtpi.org/
Paul said: “what happens if future governments and parliaments want to seriously change the modal mix in ways which reduce motor traffic volumes and hence toll income for the private operator?”
Personally, I think tolls are good, but like every other business, toll roads need to be subject to competition.
Although road users pay for their petrol, to a consumer an untolled road is free. Tolls assign a marginal cost to each trip which can have very positive effects on consumer decisions and allocative efficiency.
Toll roads with high tolls are the public transport operators dream – if governments are giving toll road operators traffic guaruntees they are stark raving mad. Telstra doesn’t have a traffic guaruntee, nor do the Ports and Airports or the privately owned electricity companies.
Joel, you’re right about tolls – they capture the externality of our contributions to everyone else’s congestion.
But I differ about toll competion. Only so many roads can serve the same market – in fact more than one is lunacy. A toll road establishes a mini-monopoly. Which makes regulation unavoidable.
If we must have regulation, then optimally regulate the network as a whole. Toll at variable times places and rates to optimally deter congestion.
Unfortunately there are all these toll roads with all these decades-long contracts with all these compensation clauses.
The Scoresby Freeway was the source of one of my favourite pieces of political nonsensespeak. When defending the his reneguing on his promise, Bracksy’s line was “The promise was made in good faith”. Broken in good faith too. My other favourite was Shadow Treasurer Simon Crean being pressed on where the GST rollback would be focused. Refusing to say he said “Our priorities are right across the board”. Great moments like these do put a bit of a spring in my step.
As AlanDownUnder points out toll roads are natural monopolies – it’s like the cops with a speed camera- good for fundraising but does stuff all about speeding.
Capturing the ‘allocative efficiency’ i.e.using tolls to manipulate traffic flows is fanciful unless the tolls are imposed over the network with the purpose of reducing/managing traffic flows as in London .
The only benefit that I can see from toll roads in PPP is that they are a government maintained trough for well connected mates. The London experience merely demonstrates the fraud that is being committed.
That last sentence should read. -The London experience merely highlights the fraud that is being committed.
Paul said: “what happens if future governments and parliaments want to seriously change the modal mix in ways which reduce motor traffic volumes and hence toll income for the private operator?�
In the case of CityLink in Melbourne – massive financial penalties against the Victorian public. Parking restrictions on side streets are stipulated in the contract, along with penalties if any public transport is built between the CBD and airport.
So thats pretty much guaranteed, Joel. And crazy, yes.
I wrote to CityLink a couple of years ago about their portrayal of people who choose not to use CityLink (and therefore drive through secondary roads and by other roundabout routes) as “rat runners”.
I objected to this because if you’re a real capitalist I think you should respect the right of the “consumer” to CHOOSE whether to “buy” your product or not. OK, I was stirring the pot here; of course Transurban/CityLink doesn’t represent classic capitalism, quite the reverse. (“Sucking on the public tit” is the expression, I believe, Mr McGuinness?) However, the spin is always that in a Market economy we should have Choice.
Therefore, to label people who chose not to “buy” their “product” as Rat Runners was a repudiation of all that the capitalist system holds dear.
I did also mention the fact that it was uncivil and insulting to people who were going about their business in a perfectly safe and legal manner.
Helen, the further irony is that roads are such a bizarre ideological mix. The free road system is just about our last socialist outpost while PPP/BOOT toll roads are commercial-in-confidence crony capitalism.
Economic rationalism would be welcome. By that I mean a market for road space. Little or no toll where road supply exceeds traffic demand and market price tolling where demand exceeds supply – market price varying depending on how much excess demand needs to be deterred wherever whenever. No spurious distinctions in the cost of road space based on when the road was constructed or how it was financed.
What really does not add up about Mitcham-Frankston and CityLink is that neither is or will be priced to internalise congestion externalities. Both are being priced on a claimed cost-recovery plus margin to operator idea.
For much of the day there is or will be no congestion on these roads so a zero price should be set. Maybe CityLink’s contract should be honoured in terms of revenue but it should be renegotiated quickly to force time-of-day pricing.
Its the same daft idea with respect to the claimed poor performance of former public mass transit. Prices for such operations should involve losses given their scale economies.
By the way the website cited by AlanDownUnder makes the standard silly claim that the external costs of congestion are measured by the costs of waiting. They are not. Some congestion is optimal given capital costs so the correct measure of congestion costs are the relevant deadweight losses not the gross waiting cost.
Harry, I am not entirely sure about this, but I think with Mitcham-Frankston the tolls are actually going to be lower during peak hour. With the grief that Bracksy was copping over having any tolls at all, this great idea contributed mightily to the winning consortium becoming the winning consortium.
Uncle Milton, If you had a reference for this it would be appreciated. It was not in the Premier’s statement. I know the Mitcham-Frankston tolls will be lower on weekends and holidays but I thought the standard rate was $4-43 for the whole 39 km trip. But if tolls are to be set lower during peak periods this would be the ultimate in daftness.
My own suggestion is to cordon price an area around Melbourne a fair bit bigger than CBD and curbside price the arterials, the Western Ring Rd and Mitcham-Frankston. Of course I favour congestion tolls that will be much lower than cost recovery tolls and which will reflect short-term marginal cost. (They would need to be a bit lower than standard first-best tolls as there are substitute roads that cannot be priced).
There need to be even tighter parking controls around the border of the cordon to stop an accumulation of externalities there and traffic calming policies on local roads to stop ‘rat running’ around the tolled arterials.
Ideally roads such as Springvale Rd should be priced but there are too many substitutes and entry/exit points so it is impractical.
And I favour dramatically lower ‘public’ transport prices for train, tram and buses with, yes, much bigger transport deficits.
I also favour greenbelts around the city that serve as amenities for urban dwellers and which create better economies for mass transit in the city periphery. One big backyard not lots of little ones! Forget about concentrating new urban developments on the fringe and servicing these with radial transport links — it won’t work since many of these outer periphery journeys are cross town. Like Mr Carr in NSW lets have integrated communities with schools and shops and planned access to transport. Barry Humphries and his silly mates are barking up the wrong tree with their criticisms of proposed self-contained communities in areas such as Camberwell.
Melbourne’s traffic problems are bad but not impossibly bad. We should get serious about piecemeal reforms until the technology arrives to comprehensively price all the major city roads electronically using GPS.
Kenneth Davidson’s piece in The Age this morning supporting a buy-back of ConnectEast from Transurban is worth comment.
The Mitcham-Frankston Rd will cost $2.5 billion to build but the Bracks government has said it would cost $7 billion to buy it back as the Liberals have suggested. What a bunch of economic geniuses these Labor politicians must be to be able to privatise a project with such a high return to the private operator!
According to Davidson however, unlike the CityLink contract there are no constraints in providing competitive services for M-F so why not encourage the Bracks Government to do this. Better bus and rail services would provide a much lower present value to Transurban and provide incentives for them to sell out at much less than $7 billion. (The figure as Davidson notes is too high anyway).
This would enable the Liberal’s buyout plan to work and save the Victorian taxpayer (and motorist) a fortune.
Of course unlike Davidson I favour levying time-of-day tolls on M-F to internalise any congestion that remains. Providing the road for free is not a good idea but neither is it a good idea to toll in a way that transfers a massive surplus to a private operator without addressing the key congestion concerns that are the guts of the reason for building this key cross-town road.
Harry, as I said earlier in this thread, the $7billion seems to be the undiscounted sum of toll receipts and other costs. This discounted sum appears to be $4.5 billion, which is still lot more than the $2.5 billion estimated cost of the building the road, but maybe the risks in the project are high. The road hasn’t been built yet, and the building costs could blow out for any number of reasons. There’s got to be some risk around the revenue forecasts too. So maybe once you take account of these risks, the return asked by Connect East isn’t too high. Or maybe it is. Only those who have seen the detailed project specs could make that judgment.
As for Davidon’s suggestion for the government to improve public transport with the explicit objective of undermining Connect East’s revenues so as to lower the buy back cost, that looks like a recipe for expensive litigation. Even if the government can get away with it, it will ensure that the private sector will never work the government again or if they did, they will demand a massive risk premium.
Davidson’s idea is probably flawed for other reasons. A lot of the traffic on the Mitcham Frankston road will be trucks and other commercial vehicles. Those drivers aren’t going to catch public transport, not matter how cheap or accessible it is. Neither will a lot of private drivers.
OK Uncle Milton I re-read more carefully.
By the way isn’t the main risk in this projects uncertainty in demands which will reflect largely innovations in transport supply. The public sector can internalise such risks but the private sector cannot which is one good reason for the public sector managing such projects — as John Q would say, it has a lower cost of capital. The risk should be borne by the sector charging the lowest price for bearing it.
Davidson claimed that the contract did not in any way inhibit the opportunity of government to make supply improvements so this type of risk is presumably built into their projected expected revenues. By the way there is one hell of a lot of cross town car traffic in Melbourne — Springvale Rd is one of the most congested Melbourne roads and it is cross town (and 20 km from the CBD!). It would partly feed Mitchum-Frankston.
I wonder about firms like Transurban. One of the biggest shareholders in Transurban is the Macquarie Infrastructure Group (MIG).
Do you recall the comments by its former Director of Infrastructure Investment, Dennis Eager? He said with respect to the M6 Midland Expressway in the UK “We can put up the tolls by whatever we like” and with respect to the Toronto Expressway “we can increase them by as much as we like”. Eager was eventually forced to resign his position.
I guess you can say that MIG are taking advantage of inept government negotiators and that it is really the contracting process that is flawed but I find it hard to lose sleep over MIG’s fate.
If Davidson’s ideas are too radical then maybe the State Government should simply try to renegotiate both the CityLink and ConnectEast contracts to leave Transurban’s bottom line undisturbed but to put into place congestion pricing.
And make sure that they exercise more care in their future contracting of such projects. Beware the wolves.
Harry, as I understand it, construction risk can be very high, such as running into rocks that are hard to move. Apparently, if they do, this can really blow out the costs. Although, such risks tend to be higher with roads that have tunnels. I don’t know if there will be any tunnels with this road.
Traffic flows, on the other hand, are fairly predctable, even with innovations in transport supply. Not that there have been many of those. We still get arounds in cars, trains and busses, as we have for the past 40 or 50 years. I don’t think it’s all that likely that we’ll all be flying around like the Jetsons any time soon.
As for the risk of the government improving alternative transport modes, yes, it probably has been priced into the contract. But that’s not the same thing as pricing in the possiblity that the government would explicitly set out to destroy Connect East’s value, which is what Davisdon seems to be suggesting.
In NSW, the M2 motorway deed had a clause dealing with subsequent construction of competing transport routes. See http://tinyurl.com/43ejd