Australian toll road lawsuit successful

Some time ago I wrote about the Clem 7 toll road in Brisbane, Australia.  It provides a north-south tunnelled bypass of the Story Bridge and other roads approaching downtown Brisbane from the south and east.

It is a PPP toll road that went bankrupt as the demand for the road was little more than half that forecast.  In December 2013, the road was taken over by Queensland Motorways, which at the time was a company owned by the Queensland State Government, but has since been privatised and is now owned by Transurban. 

The road was acquired by Queensland Motorways for A$618 million (US$472 million), but cost A$3.2 billion (US$2.4 billion) to build.

Transurban's latest traffic data reports AADT of 27,000 on the road, which is notable when you consider the forecast was that up to 100,000 a day would be using it.   

However, what was significant about this case was that around 1000 investors who bought shares in the project engaged in a class action lawsuit (PDF detailed) against the demand/revenue forecasting consultants - AECOM - and the original Rivercity Motorway company.   The Sydney Morning Herald reports that they have won a settlement gaining a payout of A$121 million (US$92 million).   The allegation was that AECOM had made traffic forecasts without reasonable grounds, and specific information had been excluded from the product disclosure (prospectus) document for investors.  

News Briefs - Australia, Canada, Ireland, South Africa, USA

Australia – Survey says distance based congestion charge would change behaviour

According to AAP, reported by Perth Now, a survey from the University of Sydney has indicated that a distance based congestion charge of A$0.05/km (US$0.08 mile) at peak times could see 22% of peak commuters driving at different times (assuming a charge between 0700-0930 and 1630-1830) and 13% to shift to public transport.

The survey comprised 1000 adults. 66% said they had no flexibility to change travel times for commuting, but the other 34% said they did.

Obviously a survey isn’t a wholly reliable measure of behaviour, but what I find telling is the potential for time shift. Far too many think congestion pricing is about mode shift, when there is as much (if not more) to gain from changing time of travel to periods when there remains spare road capacity, which means getting optimal use of the network.

Canada - British Columbia debates road pricing

According to Straight.com, British Columbian Green Party leader, Jane Sterk, has come out in favour of "pay-as-you-drive" road pricing for Vancouver, to reduce congestion and raise revenue to pay for public transport.  The same article notes that the current Minister of Transportation and Infrastructure for the province, Mary Polak (Liberal), says the issue is up to the cities to come up with a proposal and convince the provincial government that it has public support, whilst the Opposition spokesman Harry Bains prefers to consider other measures first.  The general election for British Columbia is on 14 May.

Meanwhile, the Delta Optimist has published an opinion piece by Ted Murphy who says that road pricing is likely to be the best option:

It stands to reason those who put the greatest strain on the system, and those who are most likely to benefit from any improvements, should be the ones that pay the largest share of the tab.

Conversely, it doesn't make much sense for homeowners, who are an easy mark but don't necessarily tax the transportation network, to continually be gouged every time TransLink is in need of more cash.

There's much to be worked out when it comes to road pricing, and there will undoubtedly be resistance to the idea of paying to traverse roads that up to now have been free, but at the end of the day I suspect it will be the favoured option.

It's not a question of if, but how, they're going to extract more money from us, so they might as well do it in the fairest way possible.

It isn't clear as to whether British Columbia voters think the same way.

Ireland - manual toll booths add costs to trucking firms

The Independent in Ireland reports that toll booth barriers cost them on average an extra (Euro) 0.99c each time (US$1.30) in wasted fuel.  This is with DSRC toll tags that enable automatic payment, but require trucks to slow down to a crawl to trigger the lift of the barrier.

This is crazy of course. The M50 toll road in Ireland was converted to electronic free flow a few years ago, largely because of congestion (it being the ring road for Dublin).  There ought to be a transition towards at least a mix of free flow lanes and barrier lanes.

South Africa - 24 years to repay debts for Gauteng Freeway Improvement Project

Eyewitness News reports that the South African National Roads Authority Ltd has said that it will take 24 years of toll revenues to repay the debts incurred to build the Gauteng Freeway Improvement Project. This is based on the (R)30c/km (US$0.05 per mile) rate agreed by the Government.  The maximum monthly that can be charged is R550 (around US$61).

USA - Maryland - Intercounty Connector exceeds forecasts

At a time when there are more than a few examples of toll roads that have demand well below forecasts, it is perhaps good news to report on the InterCounty Connector in Maryland (Maryland Route 200), a fully electronic toll road that opened in 2011.  According to the Washington Examiner, estimates of 30,000 daily users by June 2012 have been exceeded on the western end of the road by September 2012 (to 35,000) and not far behind on the eastern end (26,000).  The road raised US$19.7 in the year ended June 2012 compared to projections of US$18.7 million.

Brisconnections goes into receivership - another victim of toll road optimism bias?

It has now been reported by various Australian news outlets that Brisconnections, the company behind the AirportLink toll road in Brisbane, has gone into receivership following the withdrawal of support from the company's financial backers.  

Regular readers will remember my previous articles about the AirportLink toll road, which is now the second major toll road project in Brisbane to effectively go bankrupt.  They are below in chronological order, telling a sad tale of a road that faced construction cost overruns, but has been doomed by optimism bias in demand forecasting.

New Brisbane Airport Link toll road may include loyalty programme

Brisbane's latest toll road to open under clouds of scepticism

Brisbane's Airport Link toll road demand may give cause for caution

Brisbane Airport Link looking almost as bad as Clem 7

Brisbane Airport Link debacle down to incentives and value of time

It has looked increasingly doomed (at least financially) since it was opened. It is, of course, technically superb.  

Brisbane Airport Link's complex interchange at Bowen Hills

The road cost A$4.8 billion (US$5 billion) to build and has been open for seven months.  It was meant to have an average of 135,000 vehicles a day by now, but the Sydney Morning Herald said only 47,000 a day on average used it in December (a little unfair, given the Christmas season will mean numbers are down with a lack of business traffic).  Even if you take that into account, traffic numbers are around 40% of what was forecast.  ARUP, the demand modellers in this instance, will be getting legal advice and checking its insurance for fear that it may be in front of a court as AECOM now is over the Clem 7 toll road.  You can review ARUP's report to investors on its demand modelling here, on page 110.   That is worthy of a look, as you can see some of the assumptions used look now, in hindsight, to be worthy of questioning (i.e. comparing toll prices on a per km basis with other toll roads, which is not how people value time and trips, and treating historic traffic growth as reflecting future growth).

Bear in mind that the toll on the road right now is at a discount of around 45% on the full estimated price, and it still can't get the demand levels that were forecast (it is expected that the price would increase in April 2013 to about halve that discount, and increase again in October 2013, both increases testing the elasticity of demand).

The report said "a consortium of 10 banks, including ANZ and European heavyweights Deutsche Bank and BNP Paribas, finally appointed PPB Advisory as receivers on Tuesday. The banks lent about $3.3 billion to BrisConnections, and stand to lose a sizeable chunk of their money."

Of course, the road remains open.  The asset is fixed and there is nothing else that can be done with it, beyond some incremental cost trimming and some price incentives to encourage demand.  The road's true market value will be below its construction and financing costs, suggesting that it simply wasn't an economically viable project.   It would be interesting to see if there was a recent public sector comparator cost-benefit analysis to compare with the financial business case, which now looks dire.

The Courier Mail reports that Professor John Goldberg of the University of Sydney claims the modelling was used to "work backwards" from expected returns to what figures were needed, which is an outrageous assertion to make.   He has himself undertaken his own financial modelling (pdf) of toll road PPPs in Australia and is suspicious of the entire approach. However, his call for a Royal Commission to investigate this is ludicrous, as it is not a matter for the government to be specifically concerned about private companies making malinvestments and losing money.  It would be preferable for the state, if it wishes to promote specific projects as PPPs, to commission its own demand forecasts based on determining if a project is worth proceeding with.

The Product Disclosure Statement claims that debt servicing can withstand a 40% reduction in forecast traffic each year.   What happened is 60%.

What now for AirportLink?

Not much, it will continue to operate.  The receivers will commission a review of its performance to find a way to maximise revenue and minimise costs.  It will remain a prime piece of infrastructure for Brisbane.

The question is whether any legal action will occur, and what long term owner it will have.

What now for future toll road PPPs in Australia?

Disincentivising overbidding for toll road concessions

That's the title of a report for the Australian Federal Department for Infrastructure and Transport carried out by RBConsult and Oxera.

It was driven by the concerns over the list of Australian toll road concessions that appear to have been overbid with overly optimistic forecasts for traffic.  The latest being the Brisbane AirportLink, closely followed by Clem 7, the Lane Cove Tunnel and Cross City Tunnel.

I've finally had a chance to read the report, so here are some of the main points I found of interest (although it is recommended to read it thoroughly).

Six explanations were found from a literature review for possible overbidding:

- Strategic misrepresentation by the public sector:  Public sector comparators used to determine the expected costs and demand for the project get biased by political considerations demanding that the project proceed.  Bidders get caught along with this optimism.

- Strategic misrepresentation by the private sector:  Determination to because of prestige or strategic belief that winning is important in leading up to new contracts.

- Optimism Bias: A recognised flaw in forecasting of costs and benefits to take a more optimistic view of outcomes that dismisses negative influences.

- Renegotiation:  Providing an optimistic forecast to win on the basis that the contract can be renegotiated onto more favourable terms and conditions post award.

- "Winner's curse": The winner by definition will have lowered costs and have heightened expectations of demand and success, so the process self selects the one with the greatest bias from the market average.

- Technical error:  Inexperience in bidding results in bidders making a mistake that enables them to win/

It suggested various measures to minimise the risk of overbidding in future:

- In pre-procurement, subject proposed projects to economic efficiency appraisal so that only projects where there is commercial and economic viability proceed, rather than those driven by political expectations.

- Exposure to traffic risk should not be removed completely from concessions, partly to incentivise provision of higher quality service.  It is suggested that risk sharing mechanisms with the state for factors outside the control of concessionaires could be adopted, to cover factors such as GDP or population growth.   The concession should not incentivise excessive risk-taking.

- The downside risk of overbidding needs to be increased.  Options include bidding deposits that are used to cover costs if retendering is required, or to evaluate bids based on the guarantees and equity provided for the bid.

- Bid appraisal should avoid rewarding upfront premiums offered by bidders, it could also include assessment of assumptions on which demand and revenue forecasts are based.  This could also reassure investors by giving confidence that concessions are awarded to the most qualified group for appropriate reasons.

For me, the big issues are around the original viability of projects.  Governments should undertake robust commercial and economic appraisals that are allowed to say a project is not worth pursuing with.   This, at the very beginning, will mean those that proceed will be those with a reasonable chance of financial success.  An alternative would be to simply let the private sector approach government with propositions that do not involve taxpayers' money.

Beyond that is the concessioning policy framework that is set up in the first place.  It is important to note that the Queensland government has very little exposure in the Clem 7 and AirportLink toll road cases, so in that sense it hasn't been a failure.  However, if there is interest in having more of these, there needs to be a framework that allows for that.

Some ideas:

- When releasing a concession, incentivise the private sector to say if the project is not viable.  Don't make it a forgone conclusion that it will proceed.  If it is politically important then taxpayers can subsidise it, but it should be clear that bidders that are overly optimistic in forecasts wont be rescued.

- Private bidders should be required to get independent peer reviews undertaken of demand/revenue forecasts, the mere fact of this should incentivise more conservative behaviour by the primary modellers.

Beyond that, this should be more of a wake up to investors and banks that demand forecasting in this field is not something that can be done based mostly on assumptions used both on projects in other geographies and on untolled projects.  Motorists value the money they spend on tolls more than the value of time typically imputed by the public sector for transport capital projects.  It is time for investors to demand more scrutiny, and perhaps to treat the transport economics behind such forecasts as rather a bit more refined than has been undertaken by some consultants who spend most of their work dealing with untolled highway projects.

Brisbane Airport Link debacle down to incentives and valuation of time

Alan Davies in Crikey has written an excellent post on Brisbane's Airport Link toll road that specifies some of the details of the demand modelling and posits a couple of conclusions as to why it went wrong.

By now, 136,000 vehicles per day were forecast to use Airport Link at the end of the initial three month toll-free period. But on October 18 – the first day tolls were introduced – traffic was only 53,000 vehicles per day.

That's how far out estimates have been.  He notes that another set of modellers had a more sober view of demand:

Back in July, Brisbane transport modelling firm Veitch Lister Consulting released the results of (independent) modelling showing estimated traffic on Airport Link should rise to around 85,000 vehicles on a normal weekday (during school term) by April 2013. From that date the initial discounted toll will be replaced by a “medium toll”.

Veitch Lister estimates demand should drop to around 72,000 vehicles per day by October 2013 . However by November 2013 when the ramp-up to “full tolling” will be complete, the firm estimates the road might carry in the region of only 60,000 vehicles per day.

That compares with BrisConnection’s Product Disclosure Statement, which said 195,000 vehicles were expected to use the road 15 months after opening.

It's worth reminding readers that the toll in place, now, on the Airport Link is heavily discounted (49% the "full toll" level), and that for the road to be viable it needed to increase tolls to that level around a year after the road opened.  So such an independent assessment is still out, but is at least closer to reality.  Would investors have put their money into the road if those results had been in the prospectus?

That raises the first issue - which is the incentives around getting it right.  Davies suggests they are badly wrong, and one reason is that those who undertake the forecasts get paid regardless of how accurate they are.

There seems to be a systemic issue with forecasting demand for major transport projects. Maybe the incentives for over-optimism on the part of the promoters are too strong.

The way projects are put together appears to underestimate demand risk. As this press report from back in 2010 notes, “despite the failure of projects elsewhere, BrisConnections is sticking resolutely to the optimistic predictions for the Airport Link.”

Perhaps part of the problem is many of those involved in putting together major projects get most of their fees irrespective of what ultimately happens with patronage. The total underwriting and associated fees mentioned in the Airport Link Product Disclosure Statement were $89 million.

Of course, if investors demanded an independent set of forecasts, it might change things, but imagine if the whole business model for demand forecasting became incentivised around accuracy, within a range.  In other words, what if the modellers got a proportion of fees withheld for a set period, which would be paid if the forecasts were within a confidence range of say +/- 15%? This would inflate fees charged to undertake such forecasts, but wouldn't that be preferable to pouring billions of dollars into projects that are simply not viable?

The other conclusion Davies draws is that the value of time placed in transport models does not reflect what people are prepared to pay to save that time.  In other words, $3 worth of delay does not mean a motorist is willing to pay $2.50 to save that.  It just doesn't work like that.

This phenomenon isn’t new. There’s a long history of motorists, including commercial vehicle operators, going to extraordinary lengths to avoid using toll roads.

BrisConnections estimates Airport Link provides a 12 minute journey time from Bowen Hills to Toombul, compared to 25-29 minutes on alternative routes. The toll is much lower than conventional estimates of the value of the time drivers could save by using Airport Link.

On the face of it, drivers who don’t use the toll road are behaving “irrationally”. Perhaps many aren’t as good at estimating the value of their time as theory assumes. Or maybe there’s a quirk of human psychology at play – possibly many simply don’t see money and time as being readily interchangeable.

In a motoring environment where almost all trips are absent any direct pricing, the presence of a toll magnifies the cost of the toll to the driver, at least in the short term, but possibly in the medium term as well.  This boils down to the tens of thousands of motorists every day making that time/value tradeoff, suggesting the value of time of congestion isn't as high as modellers suggest.

In conclusion, this all calls for some fundamental steps that may be taken in modelling future toll roads:

1.  Get a completely independent demand/revenue modeller on top of the assigned one.  Treat it as peer review if you must, but a second opinion is looking increasingly critical.  

2.   Get both demand/revenue modellers to give confidence assessments of their forecasts.

3.  Seek demand/revenue modellers on the basis of payment for success, so that a portion of fees are retained if the forecasts prove to be accurate within a specific range.  Be careful not to incentivise overly conservative forecasts (the obvious response of modellers will be to forecast very low figures), or otherwise no project will be worth pursuing. 

4.   Use stated preference surveys to verify the value of time estimates used for forecasts. Do this every time for every new project in different locations, because it is clear that value of time is far more fluid and individual than government agencies often think they are required to assume.

Brisbane's Airport Link toll road demand may give cause for caution

The story of the bankrupt Clem 7 toll road in Brisbane, Australia is well known, and is creating shockwaves across the world of toll road concession demand forecasting.  I have also written about the attempts to make the next toll road in Brisbane - Airport Link - more successful.

According to the Herald Sun newspaper, the Airport Link toll road has been a success by some measures.  It is carrying an average of 81,500 vehicles a day, with traffic on some parallel routes dropping by up to a third, effectively eliminating congestion on those routes.  However, the forecast number is 53,300 below predictions and there is currently no toll.  

The road is currently going through a toll free promotional period, which is about to end on 17 October,   suggesting that with demand now 40% below forecasts that there are parallels with the bankrupt Clem 7 toll tunnel (although clearly nowhere near as bad), although there are expectations of demand ramping up over a 15 month period.

Understandably, Clem 7 has had a small boost in usage (17%) as Airport Link connects well with Clem 7, but is still carrying less than half the forecast volume of traffic.  

It's too early to say whether Airport Link will succeed or not, but it has to be said that when a toll road has 40% less trips that forecast - when it is free- it is difficult to be optimistic.

UPDATE:  The product disclosure statement for the project is available here.

Toll traffic and revenue studies fraught with difficulties

It's an issue that has most recently been highlighted by the court case in Australia around the bankrupt Clem 7 toll road.  It has also been highlighted by the poor performance of the Pocahontas Parkway in Virginia.  It is the accuracy and reliability of traffic and revenue forecasts for toll roads that are public-private partnerships.

The Bond Buyer has a good article summarising some of the history behind the Pocahontas Parkway and several other unfortunate projects, including comments from a range of consultants on some of the key issues and concerns.  

The Pocahontas Parkway is only meeting 60% of its predicted traffic levels, resulting in concessionaire - TransUrban - substantially writing down its value in the investment as it struggles to raise enough revenue to service the debt on the road.  

Of interest is the:

National Cooperative Highway Research Program report examining the practice of toll road revenue projections. It examined how projections of toll road revenue compared to the actual revenue obtained for 26 toll road projects between 1986 and 2004.

The report showed that only one project generated between 90% and 110% of the revenues projected within the first two years of operation. But none of the projects were on target with their projected revenues in each of the first five years of operation.

A few roads in the review, such as the GA 400 in Atlanta, were actually chronic overperformers.

What is suggests is less that there is chronic inaccuracies in forecasting, but more a systematic optimism bias in the forecasts, perhaps driven by a desire to please those touting for investment to present a positive case for the projects.   It's not about the recession either, as this study predates that.

Whilst it suggests a market failure, it is important to remember that most of the projects that get advanced are not initiated by entrepreneurs, but by governments wanting a specific project to proceed.  Of course if the project will not generate enough traffic to pay for itself, then investors need to make it clear that without government subsidies a project wont proceed.  If roads really were developed in a free market setting, then it is likely that investors would be significantly more conservative, as they would not be advancing projects for political, but rather financial reasons.

Ratings agencies are becoming sensitive to the failings of such forecasts as well:

A Standard & Poor's analyst who scored the BBB-plus Dulles Toll Road in Northern Virginia said his agency is inclined to just analyze existing historical debt without relying on a third-party revenue estimate.

Fitch analysts announced earlier this year that because of the evidence of high volatility in the earnings of managed lanes — an increasingly popular type of toll project allowing cars to pay a toll to use traditional "carpool lanes" — it will use "conservative growth estimates" for such projects in the future.

Fundamentally, new toll road projects are usually about attracting people who are using existing roads to pay more for a better trip, or to attract new trips onto a road.  If the growth in development in an area does not proceed, or motorists do not perceive enough benefit to pay for the new route it will fail.  Unless a new route offers substantial time savings for significant numbers of users, it wont generate enough revenue to pay for it.

My personal view is that Traffic and Revenue forecasts should always be peer reviewed, with the key assumptions and risks around forecasts clearly highlighted by that review and considered critically.  Investors who rarely give such forecasts much scrutiny must ask for this in the future, and take a bearish view of traffic growth in mature markets where traffic growth has tailed off in recent years.  People's willingness to pay a toll on a road has a different perceived value to their willingness to pay for other things they buy.  In addition, it should never be forgotten that trips on a road are utilitarian, they are driven by a desire to travel to a place, not to use the road per se.  Understanding the prospects for growth in those trips is critical to knowing the veracity of claims that there will be an ongoing trend of more road vehicle trips.   Is it vehicle ownership? Population growth?  Increased incomes (so increasing scope for discretionary trips)?  Tourism?  Freight?

The court case over Clem 7 is being closely watched by toll revenue forecasting consultants in many countries, and will be a key test case to see where liability lies for reliance on such forecasts and the assumptions underlying them.   Whatever the outcome, it should result in consultancies involved in this activity taking additional care to be transparent on the basis of their forecasts, but should also mean that those who commission them understand that these studies are likely to need to be longer and more intensive, if the risks and variables that affect demand are to be thoroughly evaluated.

FOOTNOTE:  The Australian Federal Government Department of Infrastructure and Transport has just released a new report called Disincentivising Overbidding for Toll Road Concessions, undertaken by Dr Robert Bain and Oxera Consulting.  I have not read it in detail, but will review it shortly.  To my knowledge it is the first substantive piece of work on this issue released by a government body (but I am happy to be corrected on it).