Brisbane's Transportation Monster

Funding the Airport Link: According to recent press reports the Airport Link project involves a 6.7km tunnel and would cost $3.8bn. Of this $1.2bn would involve equity funding (on which investors would be guaranteed a return averaging about 9% for the first 7 years) with the balance of funding from 10 banks [1, 2]

The annual revenues for such a project would need to be around $416m (assuming a 9% rate of return, a 50 year amortization period and that recurrent costs are 20% of annual capital costs). At the nominated $4.30 initial toll [1], 260,000 vehicles per day would be required to produce this return. This is more than twice the (approximately) 120,000 vehicles per day that use Brisbane's southeast freeway [1] - and is clearly unrealistic. Earlier estimates suggested that 2016 traffic volumes on the Airport Link route could be 61,000 [1]. At this volume, the toll would need to be around $18.70 per trip, which common sense suggests that few drivers would be willing to pay unless they face truly fearsome congestion on alternative routes.

In September 2010 it was reported that Airport Link project was hoped to attract 93,000 vehicles per day [1]. At this more optimistic volume the toll would 'only' need to be $12.25 per trip.

In March 2011 it was reported that new traffic forecasts suggested that the Airport Link would carry 36,000 vehicles / day more than Gateway Bridge soon after opening. 136,000 vehicles / day were expected within one month of opening in mid 2012. This is 40% above estimates submitted in 2005. Traffic was then expected to increase to 163,276 in 6 months (paying either partial toll of $3.50 or full toll of $4.80). However Matthew Burke (Griffith University) suggested that these estimates had downside risk, as freeways can only carry 1700 vehicles per hour - so a six lane road has maximum capacity of 240,000 vehicle per day - assuming it operates at peak load for 24 hours each day [1]

A complex spreadsheet and inside knowledge would be needed to model the projects revenues and operating costs reliably because:

the PDS details (a) diverse vehicle categories and several tolled sectors along with route with different projected traffic volumes that vary over time; and (b) quite complex arrangements for raising capital;
the rate of return on such projects depends on the rate at which funds are raised, whether that debt is re-rated due to monoline insurance, the effect of that debt on the company balance sheet, taxation levels and cash flows, the amount of equity raised and the amount of government contributions [personal communication];
some infrastructure companies pay distributions partly from profits and partly from debt secured on future cash flows. Brisconnections may be one of those because they are forecasting a firm initial distribution yield which indicates high confidence in their figures [personal communication];
financial viability is determined in part by what facilities promoters have available (ie if tunnelling gear is available and underused, there will be a motive to use it that is greater than if it is unavailable) [personal communication].

Firstly the global credit crunch may make it impossible in future to continue increasing debt to cover operating deficits (or even to refinance existing debts [1]). The Bank of International Settlements has realistically suggested the current destabilisation of the global economy is the product of "excessive and imprudent credit growth over a long period" [1]. Thus easy credit is unlikely to be as available in future as it has been in the past, and projects reliant on this must be suspect [1].

Secondly the core assumptions underpinning prevailing attempts to improve transport systems in Brisbane (ie the enhancement of a freeway system - of which the Airport Link project is an example) could well prove invalid, because:

Brisbane, like all Australian cities, is now subject by planning regulations to an urban 'footprint' which the balance of this document will argue, probably makes it impossible to obtain the relatively cheap rights-of-way that urban sprawl allows as the basis for freeway-based transport systems.
the global 'peak oil' event will almost certainly occur well before 2035 and presumably disrupt (and perhaps even reverse) assumptions about growth in traffic volumes which are critical to the project's apparent long term viability. Even the possible development of oil alternatives will not, unless those alternative provide cheap energy, solve this problem - and the increased fuel efficiency vehicle manufacturers are seeking [1] will at best delay it. While media commentators have suggested that the effect of high oil prices should have been analysed (rather than merely mentioned) in evaluating the Airport Link option, it is of more concern that the state government did not require such analysis before the project was approved, as the Coordinator General has apparently required that oil availability / price be considered in the environmental impact study for another tunnel option (ie the Northern Link tunnel). If a steady decline in traffic volumes is the likely outcome in a post global-peak-oil-event environment, then this will radically alter the economics of any major road project (and the commercial viability of toll roads);
passenger vehicle traffic on Brisbane roads (as measures by total distance travelled) appears to have declined at about 7% pa from 2003-04 to 2005-06 ie from 16.1bn to 13.9bn km (Apelbaum Consulting: Queensland Transport Facts - May 2008, Table 3.1-2, p50) - while travel by all road vehicles in Brisbane fell by around 5% pa (p51). The same document:
- indicates that for Queensland as a whole the total road passenger task declined by 3.3% in 2005-06 (due to both a 1.3% fall in vehicle km travelled and to a decline in vehicle occupancy - mainly in rural areas) (p 3). This was in spite of a 4.7% increase in the number of vehicles registered (p4);
- suggests that increases in oil prices accounted for a 0.8% decline in total distance travelled (p5);
- includes a diagram (Figure H4, p 6) projecting, without any obvious reason (though one can be speculated), a resumption of past rates of growth in total vehicle km in Queensland after 2008 - following the post 2003-04 stagnation;

A Conflict of Interest?

The massive longer term growth in revenues needed to make the project financially viable may or may not happen. The problem is that under current practices for private infrastructure, risk is taken by investors in securities who may not be aware of, or able to evaluate, those risks. Moreover project promoters tend to be rewarded by large initial and ongoing fees related to the capital invested rather than to the project's viability.

A recent analysis by Riskmetrics Group cast serious doubt on the sustainability of current models for private infrastructure funding from the viewpoint of investors - because: return to investors is often paid out of capital; management fees are not linked to project cash flows; aggressive approaches are taken in purchasing assets; more equity capital often needs to be raised; and debt financing is also aggressive (see Infrastructure Funds: Managing, Financing and Accounting: In Whose Interests?, April 2008). Riskmetrics suggested that private infrastructure promoters have a conflict of interest with those who invest in the securities that they promote.

A worrying parallel may well exist between the securitization of subprime mortgages by US financial institutions, whose deficiencies have been exposed, and current practices for the securitization of investment in private infrastructure assets. A desire to be able to claim a high value for their holdings of many existing similar securities might encourage promoters of new projects to present an over-optimistic view of the latter's prospects. In September 2010 it was noted that many projects funded through PPP-type arrangements had experienced financial difficulties because actual traffic volumes fell well below those that has been forecast.

Other potential (but strenuously denied) conflicts of interest have emerged in relation to the Airport Link project related to: (a) a large success fee (allegedly $500,000) for their consulting efforts paid to two prominent ex-ALP ministers; (b) the free holiday at a Sydney mansion made available to the state premier before the deal; and (c) the large purchase of BrisConnections shares by QIC (a state-owned investment company), when both have the same chairman [1]

Ongoing Congestion

Motivating drivers to pay the tolls associated with such a project requires something between a modest and a monstrous level of congestion on alternative routes. How much congestion can be quantified.

Some people would pay quite high tolls without question, as money is no object to them or their time has a large direct monetary value. Others would never pay even low tolls, as they don't have the money or don't value their time.

If an average person values their time at $10-20 (say $15) per hour, a willingness to pay a initially-nominated $4.30 toll implies an expected time saving of 18 minutes. The estimated travel time along the 6.7km Airport Link has been suggested to be six minutes (an average of 67km/ hour). The corresponding travel time along surface roads would thus need to be 24 minutes (ie 6+18) over (say) 8 km - an average of 20km / hour (cycling speed). This implies that a considerable level of surface road congestion is expected to continue to face those who choose not to pay the toll. Thus at best construction of the Airport Link is only a solution to congestion for those whose circumstances make paying the toll worthwhile.

However, if the toll were $18.70 (the level needed to cover costs without reliance on future increases in tolls and traffic volumes to cover about 75% of those costs), the required time saving would be 75 minutes and the overall travel time by surface roads 81 minutes (requiring an average speed of 6 km / hour - like that of a fast walker). This implies a truly horrendous level of ongoing congestion on surface roads.

Another Illustration of these problems: The Brisbane West Bypass

These concerns can be well illustrated by the debate about the development of a Brisbane west bypass to fulfil the same function as the Gateway Motorway does to the East. Considering such an option was a key focus of the WBTNI.

However, even before public submissions were sought on WBTNI's possible transport options, its first two options which involved western bypass routes were rejected supposedly on the basis of high costs, low traffic volumes and encouraging urban sprawl inconsistent with the SE Queensland regional plan.

The alternative appears to be a 10 km tunnel north from Toowong (WBTNI's Option 3) which, together with the Western Freeway / Centenary Highway and a new surface freeway to the north, would fulfil the necessary bypass function to Brisbane's west - though it would pass through, rather than around, Brisbane's west-side suburbs.

The trouble is that a reasonable guess-timate of the required average toll over the life of such an inner-suburban tunnel could be something like $17 per trip (assuming a capital cost of $5.7bn and average traffic volumes of 100,000 vehicles per day). As with the Airport Link this project seems to face a large financial 'black hole', and to have no prospect of 'commercial' viability without:

continued congestion on surface roads on Brisbane's west-side to force desperate drivers to pay high tolls;
a very large cash contribution by governments; and / or
investors who tolerate poorly-understood risks in purchasing infrastructure securities.

Thus the most pressing reason for the State Government's decision to reject an outer-suburban western bypass route is likely to be to concentrate and congest traffic in Brisbane's inner west-side suburbs to the potential financial benefit of the private promoters of such a tunnel (and / or to reduce necessary government cash contributions).

This might or might not be desirable, but the issues involved need to be publicly disclosed. However in making claims about low traffic volumes and high costs on an outer-suburban western bypass, there was no public disclosure about the technical basis of such conclusions. For example:

there was no way to compare the estimated cost of an outer suburban bypass with the cost of the inner-suburban alternative (a 10 km tunnel (say $5.7bn) plus the cost of upgrading existing freeways and developing others).
the transport modelling which has been used to estimate usage of alternative routes could well be unreliable. In response to an inquiry about this, the present writer received advice [1] about traffic volumes which implied that the models did not properly allow for the effect of tolls on some routes (the effect of which should be to increase the perceived 'time' involved in using a tolled route, so that the model would assign more traffic to alternative routes). If such allowances are not made, modelling would over-estimate traffic on tolled routes and under-estimate it on competing routes (such as the outer-suburban western bypass options).

Finally the fact that such a super-expensive tunnel would presumably be funded through a public private partnership gives rise to huge potential fees and thus to concerns about possible conflicts between the interests of the general community and the promoters of such projects (including public officials who may profit through a later change of career).

And Another: the Clem7 Tunnel

The funding 'black hole' that appears to affect the Airport Link project (ie a reliance for covering about 3/4 of the capital costs on: (a) borrowing to pay distributions to investors for many years; and (b) the expectation that inflation and future traffic volumes will allow costs to be recovered sometime in the long distant future) seems also to affect the Clem7 tunnel.

The annual revenue needed to cover the costs of this project (assuming construction costs (say) $3.5bn; 9% return on investment; 50 year amortization period; and 20% operating costs that are 20% of capital costs) is about $320m - ie about $0.9m per day. Traffic estimates for the tunnel are 60,000 vehicles / day [1]. With average tolls of (say) $4.50, revenue would be $0.27m / day.

Traffic volumes for this tunnel are forecast, by the project's promoters to increase to 100,000 by 2012 and 135,000 in 2026 - even though state government estimates that the total number of car trips in the region is actually likely to decline over that period

In March 2010 other analysts concluded that the Clem 7 tunnel faced difficulties - though these would be less serious than suggested above. Specifically it was noted that: (a) average initial usage during toll-free trial period was 67,000 vehicles; (b) the operators debt is $1.355bn - which requires 80,000 vehicles per day; (c) usage typically increases 40-60% after initial opinion - but toll-free traffic tends to fall 50% after tolls are applied (eg for Eastlink toll-free traffic numbers were still down 35% after 18 months); (d) this suggests that usage in 2011 could be 50,250 vehicles per day - which (at $4.43 for average vehicle) would generate $81.2m, and a net $50m after deducting $31m (PDS estimated of operating costs). If interest rates average 7.1% then interest bill would be over $90m - $40m more than operating cashflow [1]

In September 2010, it was reported that there could be a need to change the way in which infrastructure is funded because the Clem7 project: (a) incurred a huge loss and was written down by $1.56bn; (b) traffic volumes were still much below expectations despite a 50% cut in tolls; and the company might be unable to continue operating unless traffic volumes increased [1]. It was also reported that daily traffic volumes have proved much lower than forecast for freeway projects in Australia (eg Clem7 averages 20,000 rather than forecast 60,000; Land Cove Tunnel attracts 56,000 rather than 100,000; and Eastlink network has 160,000 rather than expected 258,000) [1]

This might, for example, involve:

a deliberate shift in emphasis in development of Brisbane's and Queensland's transport systems from roads / freeways to mass transit, and a corresponding shift in urban forms and in logistics;
exploration of more cost-effective options for improving the road network. For example:
- existing routes might be upgraded sufficiently to cope with less ambitious future traffic volumes;
- elevated roadways (on piers) might provide less expensive scope for new rights of way for roads than tunnelling;
- if tunnelling is the only way to get new inner-city rights of way, these should arguable be devoted to mass transit rather than motor vehicles (so as to gain economies from larger numbers of users);
- the development of an outer-suburban Brisbane west bypass could be reconsidered;
- congestion pricing, as long advocated by the RACQ, might be used to reduce demand for congested road space;
- development of transport models which explicitly consider the effect of tolls on traffic volumes - if this is not already done;
the formulation of at-least outline plans appropriate to alternative scenarios through a process which involves those with responsibilities for practical initiatives. It is understood that the Departments of Main Roads and Transport undertook an analysis of the implications of various relevant alternatives some years ago (ie of alternative urban structures in SE Queensland) and of external conditions that could have significant effects (eg constrained energy supplies or a global economic meltdown). However there seems to be no connection between such analysis in the abstract and the planning of practical initiatives - though, as noted above, constrained cheap oil supplies could radically alter the economic and commercial viability of road projects.

Examples: Selling a Lemon: The Kenmore Bypass considered the flawed investigation of a supposedly 'local' road which was being studied in isolation though:

it could be of no benefit without prior upgrading of the already-overloaded-at-peak-hour Centenary Highway / Western Freeway, and its construction would significantly increase traffic flows through Brisbane's western suburbs and require the development of other very expensive roads;
the upgrades to those other highways (which were not even being considered by the 'local' study) would not be viable because at least some of the widely-assumed tunnel-freeway connections at the northern end of the Western Freeway would probably be economically non-viable - and inconsistent with the likely need to rethink the character of SE Queensland's transport system generally.

Similarly the development of the (so called) Northern Link Tunnel (from the northern end of the Western Freeway to the Inner City Bypass) was also being investigated in isolation, through (a) it was only one of three freeway tunnels that had been envisaged for various reasons as connecting to the Western Freeway at the Toowong roundabout; and (b) it would probably be of little benefit without prior upgrading of the overloaded-at-peak-hour Inner City Bypass.

Examples: A study of the implications of the global peak oil event was apparently undertaken for the State Government, and concluded that Queensland should respond by developing various natural-gas-based substitutes (see Integrated Regional Transport Plan for SE Queensland: A Case for Increased Public and Business Awareness and Participation). However

the report did not seriously examine what such substitutes might cost, or recognise that prices in the post-peak-oil era would depend on the price of mass-produced substitutes in international markets. Moreover;
in future planning to adapting to the peak-oil event, the report proposed involving only those concerned with the supply of oil substitutes, not those (eg transport and land use authorities) whose plans would need to change if perhaps-much-higher fuel prices induced a shift away from the use of motor vehicles and a shift in desired locations to reduce the transport task.

It may be that a core of the problem in responding to the peak oil event lies in the optimistic assumption that the price of post-peak-oil fuel substitutes will be roughly the same as oil-based fuels. This is illustrates by a CSIRO-backed investigation 'Fuel For Thought' which apparently examined the possible transitional escalation in the price of fuel associated with the peak oil event on the basis of computer modelling, but did not seem to seriously study the physics, industrial chemistry and economics of alternative fuels so as to provide a realistic estimate of likely future prices in the long term. The latter are likely to be significantly (eg 3-8 times) greater than oil-based fuels - because oil reflects a huge 'subsidy' from the earth's geological history in terms of its free-energy density and convenience (see notes on Time to Reassess Land Use).

Alternative fuel costs: One source indicated that a sustained price over $US70 / barrel was needed to justify investment in "Canada's tar sands, the deep-water fields off Brazil, and Russia's "High North" above the Arctic Circle .. (as well as) biofuels from grains (sugar is cheaper)" and that bioethanol accounted for 60pc of extra oil supply worldwide from 2007-2008 and so controls the global crude price [1]. [Comment: The problem is that there is limited potential in such sources (eg biofuels conflict with food production; (b) tar sands are problematic [1]). The real question is what prices might be required to support sustained global production of (say) 50m barrels / day (or equivalent) in oil substitutes?]
Shell's CEO suggested that competition amongst energy sources (eg petrol, diesel, electricity (via hybrids), bio-fuels, natural gas and hydrogen) will drive innovation that ensures that energy remains affordable. Fuel efficiency will increase. [1]

Optimistic assumptions regarding future alternative-fuel prices is arguably the reason that officially-supported projections of Brisbane's traffic make the assumption that total vehicle traffic will resume growing at traditional rates (rather than continuing to decline as it had been observed to do since 2003) once the transitional price escalation associated with higher fuel prices ends.

Another example of critical interconnections between issues that have not been effectively managed lies in the impact of the global financial crisis on the funding of toll roads. As noted above, this had been seen to be possible by deferring much of the cost to the distant future by complex financial engineering, a practice which the credit crisis has presumably rendered permanently impractical.

Introduction Airport Link: An Example Why a Transport Problem The Monstrous 'Solution' Conclusion Progress?	Introduction Brisbane residents have become desperate for a solution to rapidly increasing traffic congestion. Unfortunately governments have been pursuing 'solutions' that simply involve expanding traditional transport arrangements - and the results are inadequate. The core of those 'solutions' involves a set of new links to fill gaps in the city's freeway network and some increased emphasis on public transport. Unfortunately: the state government's adoption of an 'urban footprint' in SE Queensland to prevent urban sprawl also denies access to the cheap 'rights of way' beyond the urban fringe that cities traditionally have had available where they focus on freeway, rather than mass transit, options; many of the envisaged freeway links thus involve tunnelling which makes them so expensive (eg $500m-750m per km and thus $2-3 per vehicle per km with 100,000 daily traffic volumes) that they are unlikely to be viable or affordable; the global peak oil event is likely to significantly affect the cost of motor vehicle transport in the reasonably near future and require major changes in urban form and transport systems; the machinery for planning the development of infrastructure in Queensland (including transport) is substandard, and subject to increasing conflicts of interest which prevent the community gaining a true picture of their problem. The 'solution' that was on offer to the Brisbane community might have been seen to be appropriate 30-40 years ago but is not a 21st century solution. A radical shift in community transportation requirements (and voter expectations) can realistically be expected over the next decade, and governments would be wise to take this into account. In March 2009 the present writer was advised by Queensland Transport of an initiative to develop an integrated and more realistic set of transport options for SE Queensland - though this arguably did not go far enough because the SE Queensland regional plan would be taken as given though it probably needed adjustment to reflect likely major constraints on future transport system options. Finally it is noted that in August 2010 a new draft regional transport plan (Connecting SEQ 2031: An Integrated Regional Transport Plan for SE Queensland) was released by the state government, though the latter is not addressed in this document. One observer suggested however that the proposal was likely to be more of a utopian dream than a practical scheme, because it did not involved costed / budgeted projects and would certainly be very expensive [1]
Airport Link: An Example + Funding the Airport Link A Conflict of Interest? Ongoing Congestion	Airport Link: An Example of the 'Monster'? The Airport Link project is one of three that have recently been announced by the Queensland Government and a private consortium as a 'solution' to problems of congestion in the vicinity of the Brisbane Airport [1, 2]. Unfortunately a rough estimate suggests that the project faces a funding 'black hole', which (as argued below) gives an insight into the deficiencies of many other current proposals (eg the Clem7 tunnel concerning which rough estimates are presented below). Funding the Airport Link: According to recent press reports the Airport Link project involves a 6.7km tunnel and would cost $3.8bn. Of this $1.2bn would involve equity funding (on which investors would be guaranteed a return averaging about 9% for the first 7 years) with the balance of funding from 10 banks [1, 2] The annual revenues for such a project would need to be around $416m (assuming a 9% rate of return, a 50 year amortization period and that recurrent costs are 20% of annual capital costs). At the nominated $4.30 initial toll [1], 260,000 vehicles per day would be required to produce this return. This is more than twice the (approximately) 120,000 vehicles per day that use Brisbane's southeast freeway [1] - and is clearly unrealistic. Earlier estimates suggested that 2016 traffic volumes on the Airport Link route could be 61,000 [1]. At this volume, the toll would need to be around $18.70 per trip, which common sense suggests that few drivers would be willing to pay unless they face truly fearsome congestion on alternative routes. In September 2010 it was reported that Airport Link project was hoped to attract 93,000 vehicles per day [1]. At this more optimistic volume the toll would 'only' need to be $12.25 per trip. In March 2011 it was reported that new traffic forecasts suggested that the Airport Link would carry 36,000 vehicles / day more than Gateway Bridge soon after opening. 136,000 vehicles / day were expected within one month of opening in mid 2012. This is 40% above estimates submitted in 2005. Traffic was then expected to increase to 163,276 in 6 months (paying either partial toll of $3.50 or full toll of $4.80). However Matthew Burke (Griffith University) suggested that these estimates had downside risk, as freeways can only carry 1700 vehicles per hour - so a six lane road has maximum capacity of 240,000 vehicle per day - assuming it operates at peak load for 24 hours each day [1] Clearly the consortium proposing to develop the Airport Link are relying on financial engineering to compensate for the initial gap between revenues and the Airport Link's capital cost. On the surface revenues would initially cover only one quarter of annual costs, and the project faces something like a $2.5bn 'black hole'. Despite the complexity of such a project, the Product Disclosure Statement (PDS) released by the BrisConnections Consortium provides some insight into how this might be achieved. A complex spreadsheet and inside knowledge would be needed to model the projects revenues and operating costs reliably because: the PDS details (a) diverse vehicle categories and several tolled sectors along with route with different projected traffic volumes that vary over time; and (b) quite complex arrangements for raising capital; the rate of return on such projects depends on the rate at which funds are raised, whether that debt is re-rated due to monoline insurance, the effect of that debt on the company balance sheet, taxation levels and cash flows, the amount of equity raised and the amount of government contributions [personal communication]; some infrastructure companies pay distributions partly from profits and partly from debt secured on future cash flows. Brisconnections may be one of those because they are forecasting a firm initial distribution yield which indicates high confidence in their figures [personal communication]; financial viability is determined in part by what facilities promoters have available (ie if tunnelling gear is available and underused, there will be a motive to use it that is greater than if it is unavailable) [personal communication]. The secret seems to be that: the cost of operations and distributions to investors will rely on increasing debt levels until around 2035, and only after this would debts start to be reduced (see Figure 9.2 in the PDS). Thus the very long term compounding effect of both inflation and projected increases in traffic volumes (about 2% pa) are being relied upon to (eventually) defray capital costs; the extremely reliable cash flows from infrastructure assets have apparently allowed: (a) a much higher level of gearing (ie 70% of capital costs has tended to be provided by banks) than is normal for corporate borrowers (ie about 40%); and (b) the small equity component to carry an apparently high rate of return [1] - if everything goes according to plan. This aggressive financial engineering would seem to create significant risks for investors in Airport Link securities. Firstly the global credit crunch may make it impossible in future to continue increasing debt to cover operating deficits (or even to refinance existing debts [1]). The Bank of International Settlements has realistically suggested the current destabilisation of the global economy is the product of "excessive and imprudent credit growth over a long period" [1]. Thus easy credit is unlikely to be as available in future as it has been in the past, and projects reliant on this must be suspect [1]. Secondly the core assumptions underpinning prevailing attempts to improve transport systems in Brisbane (ie the enhancement of a freeway system - of which the Airport Link project is an example) could well prove invalid, because: Brisbane, like all Australian cities, is now subject by planning regulations to an urban 'footprint' which the balance of this document will argue, probably makes it impossible to obtain the relatively cheap rights-of-way that urban sprawl allows as the basis for freeway-based transport systems. the global 'peak oil' event will almost certainly occur well before 2035 and presumably disrupt (and perhaps even reverse) assumptions about growth in traffic volumes which are critical to the project's apparent long term viability. Even the possible development of oil alternatives will not, unless those alternative provide cheap energy, solve this problem - and the increased fuel efficiency vehicle manufacturers are seeking [1] will at best delay it. While media commentators have suggested that the effect of high oil prices should have been analysed (rather than merely mentioned) in evaluating the Airport Link option, it is of more concern that the state government did not require such analysis before the project was approved, as the Coordinator General has apparently required that oil availability / price be considered in the environmental impact study for another tunnel option (ie the Northern Link tunnel). If a steady decline in traffic volumes is the likely outcome in a post global-peak-oil-event environment, then this will radically alter the economics of any major road project (and the commercial viability of toll roads); passenger vehicle traffic on Brisbane roads (as measures by total distance travelled) appears to have declined at about 7% pa from 2003-04 to 2005-06 ie from 16.1bn to 13.9bn km (Apelbaum Consulting: Queensland Transport Facts - May 2008, Table 3.1-2, p50) - while travel by all road vehicles in Brisbane fell by around 5% pa (p51). The same document: indicates that for Queensland as a whole the total road passenger task declined by 3.3% in 2005-06 (due to both a 1.3% fall in vehicle km travelled and to a decline in vehicle occupancy - mainly in rural areas) (p 3). This was in spite of a 4.7% increase in the number of vehicles registered (p4); suggests that increases in oil prices accounted for a 0.8% decline in total distance travelled (p5); includes a diagram (Figure H4, p 6) projecting, without any obvious reason (though one can be speculated), a resumption of past rates of growth in total vehicle km in Queensland after 2008 - following the post 2003-04 stagnation; Moreover : the way in which private infrastructure packages currently tend to be structured has caused apparently-credible analysts to fear that the promoters of such projects can profit even if investors do not (see below); Brisconnections' traffic estimates for the proposed Airport Link have been claimed to be double those by other groups who put forward proposals and by the state government [1]; there seems to be a fundamental inconsistency between projections of substantial growth in traffic volumes (which the project's financial engineering requires) and the fact that a report showed at the same time that this project was committed that road traffic in Brisbane had actually been declining. Likewise in 2010, it was reported that a suggested strategy for transport in SE Queensland (Connecting Queensland 2030) included projections of declining numbers of car trips in the region. In May 2011 it was reported that growth in car usage in all Western cities essentially ceased in 2004 [1]; the project would only be a limited solution to traffic congestion - as substantial levels of congestion must continue to affect alternative routes to force some drivers to pay significant tolls; the RACQ has argued (with unknown validity) that even modest tolls (eg $4 that could be charged for using a major tunnel if most costs are deferred by borrowing to pay early distributions) would discourage use of toll roads and force traffic back onto other heavily congested roads. They favour congestion charges as an alternative [1]. (Note: one observer suggested the costs of congestion charging could outweigh the benefits [1]) A Conflict of Interest? The massive longer term growth in revenues needed to make the project financially viable may or may not happen. The problem is that under current practices for private infrastructure, risk is taken by investors in securities who may not be aware of, or able to evaluate, those risks. Moreover project promoters tend to be rewarded by large initial and ongoing fees related to the capital invested rather than to the project's viability. A recent analysis by Riskmetrics Group cast serious doubt on the sustainability of current models for private infrastructure funding from the viewpoint of investors - because: return to investors is often paid out of capital; management fees are not linked to project cash flows; aggressive approaches are taken in purchasing assets; more equity capital often needs to be raised; and debt financing is also aggressive (see Infrastructure Funds: Managing, Financing and Accounting: In Whose Interests?, April 2008). Riskmetrics suggested that private infrastructure promoters have a conflict of interest with those who invest in the securities that they promote. A worrying parallel may well exist between the securitization of subprime mortgages by US financial institutions, whose deficiencies have been exposed, and current practices for the securitization of investment in private infrastructure assets. A desire to be able to claim a high value for their holdings of many existing similar securities might encourage promoters of new projects to present an over-optimistic view of the latter's prospects. In September 2010 it was noted that many projects funded through PPP-type arrangements had experienced financial difficulties because actual traffic volumes fell well below those that has been forecast. Other potential (but strenuously denied) conflicts of interest have emerged in relation to the Airport Link project related to: (a) a large success fee (allegedly $500,000) for their consulting efforts paid to two prominent ex-ALP ministers; (b) the free holiday at a Sydney mansion made available to the state premier before the deal; and (c) the large purchase of BrisConnections shares by QIC (a state-owned investment company), when both have the same chairman [1] Ongoing Congestion Motivating drivers to pay the tolls associated with such a project requires something between a modest and a monstrous level of congestion on alternative routes. How much congestion can be quantified. Some people would pay quite high tolls without question, as money is no object to them or their time has a large direct monetary value. Others would never pay even low tolls, as they don't have the money or don't value their time. If an average person values their time at $10-20 (say $15) per hour, a willingness to pay a initially-nominated $4.30 toll implies an expected time saving of 18 minutes. The estimated travel time along the 6.7km Airport Link has been suggested to be six minutes (an average of 67km/ hour). The corresponding travel time along surface roads would thus need to be 24 minutes (ie 6+18) over (say) 8 km - an average of 20km / hour (cycling speed). This implies that a considerable level of surface road congestion is expected to continue to face those who choose not to pay the toll. Thus at best construction of the Airport Link is only a solution to congestion for those whose circumstances make paying the toll worthwhile. However, if the toll were $18.70 (the level needed to cover costs without reliance on future increases in tolls and traffic volumes to cover about 75% of those costs), the required time saving would be 75 minutes and the overall travel time by surface roads 81 minutes (requiring an average speed of 6 km / hour - like that of a fast walker). This implies a truly horrendous level of ongoing congestion on surface roads.
Why a Transport Problem	Why Does Brisbane Have a Transport Problem? Brisbane has long suffered deficiencies in arrangements for development of its transport system (eg see Failure of Planning for Brisbane's Transport System as a Whole). The latter includes a very brief account of the history of this process, and of the growing chorus of criticism several years ago that prompted renewed efforts by all levels of government to find solutions. In the 1960s a road transport plan for Brisbane was formulated by US consultants (Wilbur Smith). This plan involved an at-the-time-progressive system of freeways and arterial roads, whose effect on encouraging urban sprawl was not really considered. Public transport options were not evaluated for another decade, and in doing so Brisbane's road systems and evolving urban form were taken as given. However the philosophies underpinning such plans were not the only source of later problems. Since the 1970s there has been both (a) an acceleration in population growth in SE Queensland and (b) a steady erosion of the capacity of governments to plan and develop infrastructure generally (as is well illustrated by other infrastructure crises that have emerged - eg those related to electricity distribution, hospitals and water supply). A general account of the history of the latter decline in institutional capacity is presented in Defects in Infrastructure Planning and Development. The latter refers, for example, to: federal fiscal imbalances which, especially since the 1970s, have (a) made it difficult for states to take real responsibility, or be held politically accountable, for their nominal functions, and (b) encouraged a shift in control from those with skills in 'doing the job' to those skilled in lobbying for funding; neglect of mainstream public administration in the 1980s, as the interests of private investors in 'major projects' were emphasised by the government of the day; failure of amateurish attempts at reform in the 1990s which (for example): eroded the knowledge and skill base in the public sector; fragmented responsibility for planning infrastructure in an attempt to promoted commercialization and corporatisation; and established centralised and regional strategic planning machinery which separated 'decision making' from those with the knowledge needed to do so competently; the complexity added by trying to promote private responsibility for functions in which effective competitive markets can not be developed. Because of such institutional defects, practical action to cope with the pressure on Brisbane's transport system from rapid population increase has been limited and the city has been suffering increasingly serious traffic congestion.
The Monstrous 'Solution + Brisbane West Bypass Clem7 Tunnel	The Monster Unfortunately, rather than a solution, Brisbane's citizens are now being presented with a 'monster' - largely because the political establishment is responding to past public expectations that existing transport systems merely need to be expanded rather than transformed. In particular public comment has been sought by the State Government on a series of future transport routes as an interim outcome of the $17m Western Brisbane Transport Network Investigation. In essence these involve: diverse proposals for cross-city freeways (often super-expensive projects through tunnels like the Airport Link mentioned above) to close gaps in the city's freeway network, combined with several new especially-outer-suburban public transport routes. One problem is the freeway tunnels seem to cost about $500m-750m per km, which (based on a parallel with the Tugun bypass [1]) seems to be about 7-10 times the cost of an equivalent surface road if rights of way are available. As illustrated by the Airport Link example above the super-expensive freeways envisaged would: produce a poor return on capital which will quickly discourage further investments, if unrealistically low tolls are claimed to secure project approval. be financially viable only if surface-level roads remain so congested that some desperate drivers pay huge tolls - if the latter are realistic. These problems would clearly be exacerbated if the global 'peak oil' event (which could well occur in the next few years) prevents the rapid growth in traffic volumes that seem to be required by the aggressive financial engineering techniques apparently envisaged in funding such projects. Another obvious problem is that the planning process does not ensure that outcomes reflect the community interest. That process is anything but transparent (see example below), and potentially involves serious conflicts of interest between the promoters of privately financed infrastructure projects and the community (in addition to those with the investors in infrastructure securities mentioned above). And unfortunately Queensland's system of government currently seems highly susceptible to unethical behaviour by state officials (eg see [1], Reform of Queensland Institutions or a Rising Tide of Public Hypocrisy?) A set of proposals which continue to focus transport on road / freeway networks might have seemed reasonable 30-40 years ago but is not appropriate in the 21st century. For example: SE Queensland's regional plan now seeks to limit the urban sprawl typically associated with a freeway-based transport system - and this seems to leave no options for freeway locations other than those which are very expensive. If urban sprawl is to be prevented there may be no practical alternative to making the painful transition to a transport system based on mass transit. Needless to say, this would also require substantial changes to land use patterns so that the high population densities required for viable mass transit emerge; a global 'peak oil' event in the planning period (given the costs, infrastructure and time required to implement currently-unplanned alternative technologies) could well force such changes in transport systems (eg steadily declining, rather than ever increasing, traffic volumes while demand for public transport rises) and urban forms, that it is irresponsible not to formally make this the subject of an alternative scenario for which a complete set of transport options would be developed. The present writer suggested the need to develop such an alternative scenario in response to 'business as usual' options put forward by the WBTNI. Independently advice [1] was received from the consultation team for the Northern Link (Option 4) that the Coordinator General required the peak oil issue to be addressed in relation to investigating that specific route. This seems like 'passing the buck' as it is surely critical in deciding the core philosophy of the transport system as a whole, and should not have to be determined in evaluating individual links. Note: It was drawn to the present writer's attention that the state government is developing an oil vulnerability strategy which is considering what would be required to mitigate the effect of peak oil. However this seems to be being done independently of the governments transport planning processes. Moreover the Brisbane City Council produced a report ('A Call for Action' respectively) which considered both climate change and peak oil issues; did not seem to deal with the peak oil challenge in much depth; produced mainly motherhood proposals; and was rejected politically. Another Illustration of these problems: The Brisbane West Bypass These concerns can be well illustrated by the debate about the development of a Brisbane west bypass to fulfil the same function as the Gateway Motorway does to the East. Considering such an option was a key focus of the WBTNI. However, even before public submissions were sought on WBTNI's possible transport options, its first two options which involved western bypass routes were rejected supposedly on the basis of high costs, low traffic volumes and encouraging urban sprawl inconsistent with the SE Queensland regional plan. The alternative appears to be a 10 km tunnel north from Toowong (WBTNI's Option 3) which, together with the Western Freeway / Centenary Highway and a new surface freeway to the north, would fulfil the necessary bypass function to Brisbane's west - though it would pass through, rather than around, Brisbane's west-side suburbs. The trouble is that a reasonable guess-timate of the required average toll over the life of such an inner-suburban tunnel could be something like $17 per trip (assuming a capital cost of $5.7bn and average traffic volumes of 100,000 vehicles per day). As with the Airport Link this project seems to face a large financial 'black hole', and to have no prospect of 'commercial' viability without: continued congestion on surface roads on Brisbane's west-side to force desperate drivers to pay high tolls; a very large cash contribution by governments; and / or investors who tolerate poorly-understood risks in purchasing infrastructure securities. Thus the most pressing reason for the State Government's decision to reject an outer-suburban western bypass route is likely to be to concentrate and congest traffic in Brisbane's inner west-side suburbs to the potential financial benefit of the private promoters of such a tunnel (and / or to reduce necessary government cash contributions). This might or might not be desirable, but the issues involved need to be publicly disclosed. However in making claims about low traffic volumes and high costs on an outer-suburban western bypass, there was no public disclosure about the technical basis of such conclusions. For example: there was no way to compare the estimated cost of an outer suburban bypass with the cost of the inner-suburban alternative (a 10 km tunnel (say $5.7bn) plus the cost of upgrading existing freeways and developing others). the transport modelling which has been used to estimate usage of alternative routes could well be unreliable. In response to an inquiry about this, the present writer received advice [1] about traffic volumes which implied that the models did not properly allow for the effect of tolls on some routes (the effect of which should be to increase the perceived 'time' involved in using a tolled route, so that the model would assign more traffic to alternative routes). If such allowances are not made, modelling would over-estimate traffic on tolled routes and under-estimate it on competing routes (such as the outer-suburban western bypass options). Finally the fact that such a super-expensive tunnel would presumably be funded through a public private partnership gives rise to huge potential fees and thus to concerns about possible conflicts between the interests of the general community and the promoters of such projects (including public officials who may profit through a later change of career). And Another: the Clem7 Tunnel The funding 'black hole' that appears to affect the Airport Link project (ie a reliance for covering about 3/4 of the capital costs on: (a) borrowing to pay distributions to investors for many years; and (b) the expectation that inflation and future traffic volumes will allow costs to be recovered sometime in the long distant future) seems also to affect the Clem7 tunnel. The annual revenue needed to cover the costs of this project (assuming construction costs (say) $3.5bn; 9% return on investment; 50 year amortization period; and 20% operating costs that are 20% of capital costs) is about $320m - ie about $0.9m per day. Traffic estimates for the tunnel are 60,000 vehicles / day [1]. With average tolls of (say) $4.50, revenue would be $0.27m / day. Traffic volumes for this tunnel are forecast, by the project's promoters to increase to 100,000 by 2012 and 135,000 in 2026 - even though state government estimates that the total number of car trips in the region is actually likely to decline over that period In March 2010 other analysts concluded that the Clem 7 tunnel faced difficulties - though these would be less serious than suggested above. Specifically it was noted that: (a) average initial usage during toll-free trial period was 67,000 vehicles; (b) the operators debt is $1.355bn - which requires 80,000 vehicles per day; (c) usage typically increases 40-60% after initial opinion - but toll-free traffic tends to fall 50% after tolls are applied (eg for Eastlink toll-free traffic numbers were still down 35% after 18 months); (d) this suggests that usage in 2011 could be 50,250 vehicles per day - which (at $4.43 for average vehicle) would generate $81.2m, and a net $50m after deducting $31m (PDS estimated of operating costs). If interest rates average 7.1% then interest bill would be over $90m - $40m more than operating cashflow [1] In September 2010, it was reported that there could be a need to change the way in which infrastructure is funded because the Clem7 project: (a) incurred a huge loss and was written down by $1.56bn; (b) traffic volumes were still much below expectations despite a 50% cut in tolls; and the company might be unable to continue operating unless traffic volumes increased [1]. It was also reported that daily traffic volumes have proved much lower than forecast for freeway projects in Australia (eg Clem7 averages 20,000 rather than forecast 60,000; Land Cove Tunnel attracts 56,000 rather than 100,000; and Eastlink network has 160,000 rather than expected 258,000) [1] There is a real risk that the general community (who are likely to pay through tolls, ongoing surface congestion, taxes and investment in risky securities) might be being misled.
Conclusion	Conclusion The road transport options that are currently being approved and considered by governments in the hope of improving Brisbane's transport system seem quite inadequate because of their: high costs; probable inability to seriously reduce congestion; apparent incompatibility with the intent of the SE Queensland regional plan; and inadequacy in the event that the global 'peak oil' event occurs in the next few years. Moreover the planning process appears to be affected by serious potential conflicts of interest because of: a lack of independent professionalism in the Public Service to provide advice and restrain self-interested behaviour; the 'revolving door' that increasingly exists between the public and private sectors; incompatibilities between the interests of the community as road users, taxpayers and investors and the interests of those who may promote private infrastructure investment in an environment in which market discipline is impractical. The planning process also seems unable to envisage 'future oriented' options (eg the likely steady decline in traffic volumes and increased community demand for public transport that is likely when the global 'peak oil' event occurs). It is probably necessary to significantly change the direction of current efforts to ease Brisbane's transport woes. This might, for example, involve: a deliberate shift in emphasis in development of Brisbane's and Queensland's transport systems from roads / freeways to mass transit, and a corresponding shift in urban forms and in logistics; exploration of more cost-effective options for improving the road network. For example: existing routes might be upgraded sufficiently to cope with less ambitious future traffic volumes; elevated roadways (on piers) might provide less expensive scope for new rights of way for roads than tunnelling; if tunnelling is the only way to get new inner-city rights of way, these should arguable be devoted to mass transit rather than motor vehicles (so as to gain economies from larger numbers of users); the development of an outer-suburban Brisbane west bypass could be reconsidered; congestion pricing, as long advocated by the RACQ, might be used to reduce demand for congested road space; development of transport models which explicitly consider the effect of tolls on traffic volumes - if this is not already done; the formulation of at-least outline plans appropriate to alternative scenarios through a process which involves those with responsibilities for practical initiatives. It is understood that the Departments of Main Roads and Transport undertook an analysis of the implications of various relevant alternatives some years ago (ie of alternative urban structures in SE Queensland) and of external conditions that could have significant effects (eg constrained energy supplies or a global economic meltdown). However there seems to be no connection between such analysis in the abstract and the planning of practical initiatives - though, as noted above, constrained cheap oil supplies could radically alter the economic and commercial viability of road projects. It is also vital to rebuild institutional capacity to: support more a more appropriate process for planning and development of infrastructure; and reduce the incidence of similar problems in future. At the very least a high level of public disclosure of technical data before, not after, firm commitments are made is vital to reduce the risk that decisions are distorted by conflicts of interest. Other reforms that seem necessary are outlined in Queensland's Next Successful Premier. This refers, for example, to the need for: the community to support institutions which provide more up-to-date and practical inputs to the political process (to overcome the chronic policy weakness of the state's political system, and allow the community to better understand their real options in dealing with complex technical issues); and a truly (rather than nominally) professional Public Service to provide practical support to the community's elected representatives and an informed watch-dog role in relation to conflicts of interest.
Progress?	Progress? In March 2009 the present writer was advised by Queensland Transport of an initiative to develop an integrated and more realistic transport plan for SE Queensland. This involved a joint Queensland Transport / Department of Main Roads (DMR) project ('Connection SEQ 2031'), which: is to complement the SEQ Regional Plan; will seek to identify transport options that meet challenges such as: reducing greenhouse gas emissions; oil supply / price constraints; lifestyle and economic impacts; and higher construction costs; recognises the inadequacy of simply providing more roads, especially when taking account of oil depletion; is likely to suggest some toll roads; will seek a transport system to suit the region's future needs. This initiative seems to address significant and long-standing concerns. Moreover, by seeking an integrated plan it should reduce obvious problems that arise in trying to develop plans for individual transport elements in isolation which can result in proposals that simply do not fit into the regional framework. Examples: Selling a Lemon: The Kenmore Bypass considered the flawed investigation of a supposedly 'local' road which was being studied in isolation though: it could be of no benefit without prior upgrading of the already-overloaded-at-peak-hour Centenary Highway / Western Freeway, and its construction would significantly increase traffic flows through Brisbane's western suburbs and require the development of other very expensive roads; the upgrades to those other highways (which were not even being considered by the 'local' study) would not be viable because at least some of the widely-assumed tunnel-freeway connections at the northern end of the Western Freeway would probably be economically non-viable - and inconsistent with the likely need to rethink the character of SE Queensland's transport system generally. Similarly the development of the (so called) Northern Link Tunnel (from the northern end of the Western Freeway to the Inner City Bypass) was also being investigated in isolation, through (a) it was only one of three freeway tunnels that had been envisaged for various reasons as connecting to the Western Freeway at the Toowong roundabout; and (b) it would probably be of little benefit without prior upgrading of the overloaded-at-peak-hour Inner City Bypass. However even the 'Connection SEQ 2031' proposal did not seem adequate because it was apparently intended to take the existing SE Queensland Regional Plan as a given in defining an integrated transport plan. The problem is that the factors that now required changes in transport systems (especially the likely effect of the global peak oil event on fuel prices) would also significantly change sensible decisions about the location of residential, commercial and industrial functions within Brisbane city and SE Queensland (eg see Time to Reassess Land Use) Fragmented and narrowly conceived investigations again seemed to be at fault. Examples: A study of the implications of the global peak oil event was apparently undertaken for the State Government, and concluded that Queensland should respond by developing various natural-gas-based substitutes (see Integrated Regional Transport Plan for SE Queensland: A Case for Increased Public and Business Awareness and Participation). However the report did not seriously examine what such substitutes might cost, or recognise that prices in the post-peak-oil era would depend on the price of mass-produced substitutes in international markets. Moreover; in future planning to adapting to the peak-oil event, the report proposed involving only those concerned with the supply of oil substitutes, not those (eg transport and land use authorities) whose plans would need to change if perhaps-much-higher fuel prices induced a shift away from the use of motor vehicles and a shift in desired locations to reduce the transport task. It may be that a core of the problem in responding to the peak oil event lies in the optimistic assumption that the price of post-peak-oil fuel substitutes will be roughly the same as oil-based fuels. This is illustrates by a CSIRO-backed investigation 'Fuel For Thought' which apparently examined the possible transitional escalation in the price of fuel associated with the peak oil event on the basis of computer modelling, but did not seem to seriously study the physics, industrial chemistry and economics of alternative fuels so as to provide a realistic estimate of likely future prices in the long term. The latter are likely to be significantly (eg 3-8 times) greater than oil-based fuels - because oil reflects a huge 'subsidy' from the earth's geological history in terms of its free-energy density and convenience (see notes on Time to Reassess Land Use). Alternative fuel costs: One source indicated that a sustained price over $US70 / barrel was needed to justify investment in "Canada's tar sands, the deep-water fields off Brazil, and Russia's "High North" above the Arctic Circle .. (as well as) biofuels from grains (sugar is cheaper)" and that bioethanol accounted for 60pc of extra oil supply worldwide from 2007-2008 and so controls the global crude price [1]. [Comment: The problem is that there is limited potential in such sources (eg biofuels conflict with food production; (b) tar sands are problematic [1]). The real question is what prices might be required to support sustained global production of (say) 50m barrels / day (or equivalent) in oil substitutes?] Shell's CEO suggested that competition amongst energy sources (eg petrol, diesel, electricity (via hybrids), bio-fuels, natural gas and hydrogen) will drive innovation that ensures that energy remains affordable. Fuel efficiency will increase. [1] Optimistic assumptions regarding future alternative-fuel prices is arguably the reason that officially-supported projections of Brisbane's traffic make the assumption that total vehicle traffic will resume growing at traditional rates (rather than continuing to decline as it had been observed to do since 2003) once the transitional price escalation associated with higher fuel prices ends. Another example of critical interconnections between issues that have not been effectively managed lies in the impact of the global financial crisis on the funding of toll roads. As noted above, this had been seen to be possible by deferring much of the cost to the distant future by complex financial engineering, a practice which the credit crisis has presumably rendered permanently impractical. This (ie the fact that interconnections amongst various policy areas make it unrealistic to deal with traditionally 'separate' issues separately) is arguably an example of the problem of complexity that is now one of the circumstances requiring an enhancement of Australia's traditional governance systems (see Australia's Governance Crisis). Restoring 'Faith in Politics' suggests the need to enhance cross-disciplinary communication through both: (a) purposeful / cohesive public services; and (b) well-developed networks amongst civil institutions that provide raw material for public policy debates.

Brisbane's Transportation Monster (2008+)
CPDS Home Contact	Debating Regional Transportation Proposals Crony Capitalism in Queensland? Solving SE Queensland's Transport Planning Woes? Tunnel Freeways: Where is the Strategic Insight? Will Three New Cites Solve SE Queensland's Urban and Transport Challenge?