A week or two back I took a trip through Brisbane’s new bright, shiny transport infrastructure project, the Clem7 tunnel. This journey was not by choice but misadventure. I was on my way from the Sunshine Coast to meet my kids in Paddington for dinner.

You may wonder how Clem7 featured in my journey from the Bruce Highway to Paddington. This is how it happened. I successfully negotiated Gympie and Lutwyche Roads through the forest of cranes, fleets of huge trucks and other assorted construction hazards seemingly stretching from Kedron to the RBH. The Inner City Bypass was my favoured route to Paddington so I kept an eye out for ICB signs. Suddenly my peripheral vision picked up a sign labelled Ipswich and the Gold Coast. Surely this must be the ICB entry so I swung left, but it didn’t seem very familiar. I caught a glimpse of a slip-road snaking up a ramp as I realised with horror that I was about to enter the yawning jaws of Clem7. A “No tag, no worries” sign greeted me as I frantically tried to find a way to turn back.

Since a tunnel trip was now inevitable I decided to sit back and enjoy it. And as road tunnels go, Clem7 is fine. It’s well lit, without any sharp twists and turns and has cheery signs to read every few hundred metres. No evidence of the fumes reported in the press recently. Mind you, there was no other traffic to speak of, so maybe it wasn’t a comprehensive test of the extraction system. Cruising at 80km/h I eventually emerged on Ipswich Road just before the Prince of Wales Hospital. I was on the wrong side of the city for my dinner with the kids, but happy to have sampled Brisbane’s first trans-river tunnel. Rather than pay another toll I headed back to Paddington across the Story Bridge. It didn’t seem to take much longer than the tunnel trip. There were a few traffic lights to negotiate but out of peak hours it was a reasonably quick journey. Even with the detour I made dinner on time.

The next morning I wondered how many motorists would choose Clem7 and pay $2.95 to save a few minutes travel time. A little research revealed not nearly enough. The River City Motorway Group, Clem7 owners and operators, will be facing financial melt-down if patronage does not improve dramatically over the next year or so. At the moment fewer than 30,000 vehicles a day use the tunnel, not much more than a third of the number originally predicted at this point in Clem7’s early life. And now the 30% discounted toll period has been replaced in desperation with an even lower $2 toll, so present revenues must be less than 20% of the initial projections. The original forecasts suggested traffic volumes of 90,000 within the next six months and over 100,000 trips per day by the middle of next year. The omens are not good and these numbers now look like pie in the sky.

Why are these traffic forecasts so wildly wrong? Surely, the Brisbane City Council, which planned and promoted the project and the directors and management of the River City Motorway Group, must have taken the best professional advice available to predict traffic volumes and revenues. These are vital elements of a business case to test the viability of a $2 billion project investment. Indeed they did. Both the BCC and the RCMG used the same well-respected engineering consultant to conduct the traffic forecasts. And in the time-honoured practice of all traffic engineers, the consultant utilised a sophisticated computer-based “Traffic Model” to develop the long-term predictions. So why did this process predict initial traffic volumes so wildly at odds with actuality?

In my view there are two basic reasons. The first is that any computer model is only as good as the information it receives. The beast needs feeding with a bucketload of assumptions on future trends and conditions to function in predictive mode. The type of traffic model used by the consultant is reasonably good at predicting broad-brush, year-by-year traffic volumes in and around Brisbane. However, an accurate prediction of the numbers of vehicles using a single travel corridor is much more difficult. Add a completely new piece of infrastructure and the equation becomes even more complicated. The human element comes into play when drivers armed with a basic knowledge of the road network are faced with congestion and changing traffic conditions. Brisbane motorists could well make route choices in the future that throw computer-based, corridor-specific predictions into disarray.

The second reason is easier to understand. This new piece of infrastructure has a direct cost attached to it, the toll. So our average car commuter is faced with a clear alternative, a free drive to work or spend $4.28 (or maybe $42.80 a week) to save, at best, 10 minutes a trip. Outside peak hours that time saving is only about five minutes. Transport modes, facilities, customer travel choices and behaviour are subject to the same laws of economics, supply and demand, market share and competition as any other commodity. Conventional computer-based traffic models are notoriously unreliable at predicting travel corridor choices involving cost and value judgements. Skilled market research and econometrics are much better tools but few engineering consultants are practiced in these disciplines.

Clem7 is a new river crossing but largely mirrors an existing crossing, the Story Bridge. A little further removed is the Captain Cook Bridge. These are both major competitors for the new tunnel. In September 2009 these two river crossings averaged 94,000 and 128,000 trips a day respectively. So the combined daily north-south city river crossings were 222,000 before Clem7 opened. Our consultant’s traffic model predicted that by late 2010, Clem7 would attract 90,000 trips a day. Where were all these predicted trips coming from? A 40% increase of cross-river traffic in one year seems highly unlikely, just as unlikely as Clem7, even with its $2 toll, capturing 40% of the market share from free trips across the other two bridge crossings. Even Blind Freddy should be able to see the basic reason for Clem7’s poor performance. Not looking good!

Another question nagging at the back of my mind is why was the toll initially set at $4.28? This seems a pretty high price to pay for a time-saving of five or, at best, ten minutes. The Gateway Bridge toll is currently $2.95. This toll bridge is much closer to a monopoly since all the competing river crossings are in the city. To avoid the toll entails a detour of several kilometres through city traffic, probably adding at least 20-30 minutes to a trip from the Gold Coast to, say, Brisbane Airport. I think most motorists would consider $2.95 a fair price to pay for this saving in time and frustration.

The River City Motorway Product Disclosure Statement, the original offer document for the equity placement, does not explain the methodology for determining the toll structure. The only clue to this pricing decision seems to be a statement made by the RCMG chairman, Bob Morris, at the 2009 annual general meeting. He stated that “motorists using our tunnel will also be paying almost 50% less on a per kilometre basis than motorists using comparable tunnels in Sydney and Melbourne”. How are these tunnels in other cities in any way comparable with Clem7? They are in completely different markets with their own unique set of competitive dynamics.

Heaven forbid, but did the promoters of this mega-project with a $2 billion price tag decided on the revenue required to give an adequate rate of return and then divided it by the number of tunnel trips predicted to come up with the toll pricing structure? It is interesting to note that one of the experienced toll-road operators originally invited to bid for the project actually declined. I wonder why? Maybe they were not convinced it was a viable project. Time will tell.

What does the future hold for the holders of RCMG stock? They have already seen their initial investment plummet from the $1 issue price to less than four cents a share. And who are these unlucky investors? A major component of the shareholder register appears to feature investment vehicles for major fund management groups. This suggests that quite possibly you and I own shares in RCMG indirectly through our pension funds. And what happens if the RCMG runs out of cash and can’t meet interest payments for its $1.3 billion in term loans or fails the debt service coverage test in 2012? Default on the loan covenants will mean the banks that provided the loans may well exercise the right to take control of the assets. In that event the poor old shareholders (and maybe our pension funds) will be left with nothing. There is not much light at the end of this tunnel.

Can we learn anything from this debacle? You could point an accusing finger at the consultants’ traffic predictions. To give them their due, the consultant engineers did state that, apart from the assumptions fed into the computer model, “future traffic volumes also depend on drivers’ willingness to pay tolls and whether the benefits offered by the toll roads (including travel time savings) are worth the payment”. They also updated a behavioural survey conducted by BCC on route choices by potential tunnel users. How this research was conducted is unclear. This type of survey has to be designed skilfully and the results interpreted with some care. Many respondents when faced with questions on hypothetical future travel options will say they will use new infrastructure if it appears to offer some time saving or other travel benefits. On project completion when faced with a real choice (and cost) they may behave quite differently. The bottom line is that the promoters of major transport infrastructure projects, in both government and the private sector, need to take a hardheaded look at viability based on pragmatic, skilled market research and economic common sense.

Campbell Newman was swept into city hall riding a tsunami of voter concern at Brisbane’s burgeoning traffic gridlock. His vision of swift delivery of major transport infrastructure improvements struck a chord with the electorate. Newman also assured us that these mega-projects with mega-price tags would not cost the Brisbane rate-payers dear. The catch phrase was Public-Private Participation (PPP) where government invite tenders from private sector companies or consortia to build own and operate public service infrastructure for a predetermined period, normally several decades. The successful bidder shoulders the majority of the project costs and associated risks. The rewards are the long-term income stream from tolls and/or fees levied on the users of the infrastructure. Clem7 is the first Brisbane PPP road tunnel to open. Jim Soorley initiated the project but Newman enthusiastically adopted it when he became lord mayor. It could serve as a bellwether for a swathe of similar initiatives already in train.

As an ex-army engineer, Newman may have taken a leaf out of the book of those intrepid Diggers who tunnelled under Hill 60 during the First World War to blow up the Boche. His transport initiatives bear a similar hallmark. He has enthusiastically promoted tunnels as the main line of attack in the war against the tyranny of Brisbane’s growing traffic congestion. Several of these future high-cost initiatives are billed as PPP projects. Let’s hope Newman succeeds and, as opposed to our Hill 60 heroes, his tunnel vision doesn’t go up in smoke.