The focus of our audit

The Cross City Tunnel (CCT) has been marked by controversy since its opening. About 30,000 cars a day are using the tunnel so far, about one-third of the expected traffic volume. The road changes have resulted in slow, tangled traffic.

There has also been substantial public concern since the CCT opened over whether: road closures and changes were necessary; the toll is too high; the contract was awarded to CrossCity Motorway Pty Ltd (CCM) fairly.

This audit is one of several reviews and inquiries into the CCT with each review addressing various aspects of the project. In this report we have focused on three key issues:

§        was the upfront payment a legitimate reimbursement of necessary expenditure?

§        were the variations to contract in the amending deed of December 2004 reasonable, and were they handled appropriately?

§        were the changes to surface roads based on a robust assessment against stated objectives?

Audit opinion

In our opinion the Government’s ‘no net cost to government’ requirement was a legitimate (but not the only possible) basis for the tunnel bid process. The Government was entitled to decide that tunnel users meet the tunnel costs.

Structuring the bid process on the basis of an upfront reimbursement of costs incurred (or to be incurred) by the Roads and Traffic Authority (RTA) was therefore appropriate.

Whether these costs should have included ‘public domain’ costs not relating directly to the tunnel (and hence passing those costs on to tunnel users) was a decision the Government was entitled to make. A core and common role of government is to redistribute income between different groups in the community.

In our opinion, however, the Government, Treasury and the RTA did not sufficiently consider the implications of an upfront payment involving more than simple project cost reimbursement (i.e. the ‘Business Consideration Fee’ component).

While the RTA may have genuinely believed it was in the public interest to ‘capture’ some of the proponent’s ‘surplus’ through the Business Consideration Fee, no real thought was given to:

§        foregoing the fee to reduce the toll charges on tunnel users, or

§        how this amount was to be used (e.g. retained by the RTA for use on other roadworks or used by Treasury for allocation to other areas of government).

In one sense this issue ultimately became academic. The RTA’s
project-related costs will eventually absorb all of the upfront payment, once all claims are settled. No Business Consideration Fee will remain.

However, the principle of requiring a Business Consideration Fee remains an issue to be resolved.

In our opinion such a fee would need strong justification where:

§         the cost of it is met, in effect, by people not party to the contract (in this case the tunnel users), or

§         it achieves a current benefit to be paid for by future tunnel users (i.e. it distorts inter-generational equity).

In our opinion the RTA was wrong to change the toll escalation factor late in 2002 to compensate the tunnel operator, CCM, for additional costs.

This action distorts inter-generational equity between tunnel users. If it was appropriate for tunnel users to fund these costs, this should have been done by changing the base tolls. Escalation factors should do no more than reflect underlying cost movements or inflation.

In our opinion the variations in the amending deed were reasonable.

The net present value of the 15 cents toll increase accurately reflected the $35 million incurred by CCM in doing the extra work required after the signing of the original contract. But it made an already expensive toll even more expensive. By 2018, the toll will be about 35 per cent higher than it would have been due to this increase and the change to the escalation formula.

The handling of the amending deed also lacked transparency:

§        it was not made public until late 2005

§        no clear breakdown of the costs has been made available publicly, even to date

§        it was done without full probity assurance.

A widely held view is that the road changes were not necessary, but were introduced to force motorists into the tunnel to profit the tunnel operator.

When the project was put to tender, only limited road changes were specified. All bids were made on this basis.

In our opinion the RTA and the Department of Planning (DoP) communications and community consultation over road changes were sound at the detailed level but not effective in conveying the overall impact of the package of changes. We cannot say that the road changes were robustly assessed, either collectively or on a road-by-road basis.

Consultation with stakeholders about the road changes failed to clarify the cumulative impact of the changes, especially in eastern Sydney. It was not inclusive enough to capture the significant resentment the changes caused. Any loss of patronage from this resentment will further hinder the tunnel’s main objective of reducing traffic in the City.

In our opinion the concept behind the road changes was to implement long-standing Government planning objectives to reduce congestion in and around Central Sydney, and to improve public transport routes and urban amenity.

The initial strategy was to make car travel on surface roads ‘unattractive’ and could therefore be described as ‘funnelling’ traffic into the tunnel. But the motivation was primarily to clear up the congestion on surface roads rather than to make the tunnel profitable. The financial viability of the tunnel, and the RTA’s interpretation of ‘no net cost to government’, did however influence some important planning decisions.

Maintaining direct toll-free alternative routes was a key principle in the DoP Director-General’s requirements, but was lost as the project developed.

Summary of recommendations

We recommend that the RTA:

§        define project costs and development costs more clearly (page 38)

§        require transport consultants to provide a clearly defined range of likely patronage outcomes (page 62)

§        consider the impact of various patronage outcomes on a project’s viability (page 62)

§        assess the affordability and public acceptability of any proposed tolls in future projects (page 47).

We recommend that Treasury and the RTA:

§        review the bidding model used for Public Private Partnership projects, including the CCT (page 33)

§        limit the upfront payment sought from the private sector to recovery of development costs, and abandon the option of a Business Consideration Fee (page 33)

§        consider receiving upfront payments progressively as project development costs are incurred (page 38)

§        consider making the toll level the point of competition in the bidding process (page 33)

§        make value for money for motorists an explicit objective of the assessment of bids for future tollway projects (page 33)

§        consider alternative funding methods for subsequent project cost increases, and ensure decisions are made with regard to motorists price sensitivity (page 38)

§        develop guidelines for setting any future tolls equitably, related to distance travelled and the cost of the project (page 51).

We recommend that Treasury and the Budget Committee of Cabinet:

§        clarify for any future contracts what ‘no net cost to government’ means, including whether agencies should use their capital budgets to cover any cost increases (page 36)

§        develop guidelines for the use of any surplus from current or future upfront payments (page 33)

§        consider direct funding of public domain improvements (page 38).

We recommend that Treasury and Premier’s Department require agencies to:

§        keep the full tender evaluation and review panels involved in complex high risk projects until the project deed is signed, and re-convene them if amending deeds are needed (page 52)

§        make any contract amendments subject to the same level of probity checks and scrutiny as the original contract process (page 52)

§        make any contract amendments, and their summaries, public in a timely manner (page 52).

We recommend that the DoP and the RTA:

§        improve the consultation process for major projects (page 70)

§        conduct a joint review of road changes that are not consistent with current traffic volumes (page 72)

§        resolve the inconsistency between current traffic arrangements and the stated objective of maintaining at least one direct toll-free alternative route on all sectors affected by the CCT (page 72).

We recommend that the DoP, in conjunction with Treasury:

§        review the use of open-ended conditions of approval of projects (page 38).

Audit findings

Chapter 2

Was the upfront payment a legitimate reimbursement of expenditure?

CCM offered the RTA $100.1 million as an upfront payment for the CCT project. The offer comprised a Development Fee of $54 million for the RTA’s initial estimate of its project development costs, and a Business Consideration Fee of $46.1 million. The actual payment reduced to just under $97 million on finalisation of the contract, mainly due to movements in interest rates.

The Business Consideration Fee was the price a proponent was willing to pay for the right to build and operate the tunnel. It was distinct from the estimated development costs. There were no government directions on its use.

The bidding model developed by Treasury allowed the bidders to compete on a number of variables. They could bid on the upfront payment, the concession period or the toll. CCM offered the highest upfront payment, and this became a decisive criterion in the assessment of bids.

The advantage of making the upfront payment a point of competition is that it makes it easier to compare bids. Difficulty in comparing bids has been the subject of past criticism. However, allowing the upfront payment to be a point of competition also presented significant risks, particularly the risk of higher tolls. If motorists perceive the level of tolls as unreasonable, they may avoid using the tunnel. Therefore, we consider that the Budget Committee of Cabinet should have reviewed and assessed the implications of this approach.

All of CCM’s bids predicted a higher number of cars using the tunnel than the RTA and the rival bidders. CCM’s projections for 2006 ranged from 12 per cent to 50 per cent higher than other bidders. This should have concerned the evaluation panel sufficiently to cause greater scrutiny of the projections. Actual patronage to date has been under a third of the CCM estimates.

After receiving the $97 million upfront payment in December 2002, the RTA used part of it to recover its project development costs, and committed another part to future CCT project costs. By April 2005, the RTA had not spent or committed about $9.0 million of the upfront payment, earmarked for the finalisation of outstanding project costs. The RTA used only $3.1 million of the unspent portion of the upfront payment on the subsequent cost increases, which largely flowed from the Planning Minister’s Project Conditions of Approval.

To recover subsequent cost increases, preserve the upfront payment, and avoid using funds from its capital expenditure budget, the RTA negotiated two deals with CCM. The first deal resulted in a change to the way the toll escalates. This was incorporated in the original contract. The second deal resulted in a 15 cents toll increase. This was incorporated in the First Amendment Deed.

CCM delivered the project to the RTA on time. The RTA delivered the project at ‘no net cost to government’ because it passed on all increased costs to motorists by way of toll increases. This was because it interpreted ‘no net cost to government’ to mean no net cost to the RTA budget. The RTA did not adequately consider using its capital expenditure budget to meet cost increases.

Chapter 3

Were the contract variations in the amending deed appropriate?

The $38.1 million cost increases identified at the amending deed stage were legitimate and were the responsibility of the RTA, not CCM.

After the contract was signed, new costs arose, totalling $38.1 million. The RTA accepted responsibility for these costs. It funded $3.1 million from the upfront payment, and CCM carried out the remaining $35 million worth of work. The RTA compensated CCM by allowing an increase of 15 cents in the base toll of $2.50 (in 1999 prices). The change was formalised in the First Amendment Deed (FAD), signed in December 2004. This is the only variation to the original contract.

The 15 cents base toll increase was an appropriate amount to reflect the net present value of the $35 million of work that CCM undertook. It will result in a 5.6 per cent increase in toll revenues to CCM.

The 15 cents base toll increase and the change to the toll escalation formula have a major and continuing impact on the toll:

§        the 15 cents increased the base toll for the main tunnel by 6.0 per cent, and for the shorter run (vehicles from the east exiting at Sir John Young Crescent) by 13.6 per cent. The RTA did not apply a pro rata increase

§        the change to the escalation formula has the biggest influence on the tolls; an increase of around $1.12 for the main tunnel by 2018. Adding the 15 cents to the base toll brings that up to around $1.43. Together, this means the toll would be 35 per cent higher than originally planned by 2018

§        the effect of the two changes is more severe for the shorter run, with an increase in the planned toll (base toll plus CPI) of 44 per cent by 2018.

There was a total of $110 million of extra project costs ($75 million at the Supplementary EIS stage resulting in changes to the toll escalation, and $35 million resulting in the 15 cents increase to the base toll). These were separate and additional to the upfront payment. If the Government had contributed this $110 million rather than pass it on to the users, the tolls could have been 51 cents lower on tunnel opening and more than one third lower by 2018.

The RTA obtained proper approval for the amending deed and instituted procedures to manage the works covered in the deed. But the handling of the amending deed is open to criticism:

§        it was not made public, amidst lack of clarity about whether the RTA was required to publish a summary, until late 2005

§        there is still no clear breakdown of the costs available publicly

§        it was entered into after all the protective structures set up to ensure fairness around the contracting process had been dismantled

§        the Treasurer’s consent to the amendment was sought in a short timeframe, which may have worked against full analysis of the issues.

Chapter 4

Were the road changes based on a robust assessment?

The RTA developed 73 road changes to integrate the CCT with the road network. Most are not road closures as such, but changes to lanes and road use. By May 2006, 63 had been completed, six reversed and four were still pending. Of the 73 road changes:

§        28 were in response to the EIS and 45 to the Supplementary EIS

§        22 are specified in the contract as Materially Adverse Events (MAEs) and so may trigger compensation to CCM if reversed.

There is uncertainty over whether any reversal of the other 51 road changes not specified as MAEs would trigger compensation.

A widely held view is that the road changes were not necessary, but were introduced to force motorists into the tunnel in order to profit the tunnel operator.

In our view this was not the case. We found that the objective of the road changes was to reduce through traffic in and around Central Sydney and to improve the public domain, not to be a tunnel funnel.

Both the RTA and the DoP relied on patronage estimates of over 80,000 vehicles a day using the tunnel in their decision to implement immediate road changes. This was because the agencies believed that such a large reduction in above ground traffic would immediately attract more cars and lose the benefits of reduced congestion. Actual usage has been below 25,000 vehicles a day (when the full toll has applied). Since the half-price period began, usage reached an average of 34,000 vehicles a day.

Maintaining toll-free alternative routes was a key principle in the DoP Director-General’s requirements. But road restrictions were added as the project developed because there was no mechanism to judge the cumulative magnitude of the road changes. This key principle was lost as the project progressed.

We cannot say that the road changes were robustly assessed, either collectively or on a road-by-road basis because:

§        the patronage scenario was not robustly assessed

§        ensuring the financial viability of the tunnel, and the RTA’s interpretation of ‘no net cost to government’, affected important planning decisions.

There was extensive consultation with stakeholders about the road changes. It did not however capture the significant resentment among prospective toll payers. This is a group which was admittedly diffuse and difficult to survey. Any loss of patronage from this resentment will further hinder the tunnel’s main objective of reducing through traffic in the City.

Response from the Roads and Traffic Authority

Response from Treasury

Thank you for providing me with a copy of the final draft of the Performance Audit – The Cross City Tunnel Project and inviting me to comment on the report.

A number of the recommendations you make are generally consistent with the Premier’s Department recent Review of Future Provision of Motorways in NSW and the Joint Select Committee’s Inquiry into the Cross-City Tunnel. As such Treasury is already in the process of implementing changes to address some of your recommendations.

In particular, your recommendation that Treasury publicly disclose contract amendments is being implemented by revising Ministerial Memorandum 2000-11 Disclosure on Information on Government Contracts with the Private Sector.

Also, Treasury is currently updating and revising the Working With Government Guidelines for Privately Financed Projects to address a number of your recommendations. In particular, the revised version will emphasise the importance that any user charges and/or taxpayer contributions for projects should be value for money and Budget Committee approval will be required prior to amending any contract.

I would also like to take the opportunity to clarify and elaborate on a few areas that you mentioned in your report.

Firstly, I would like to clarify your suggestion that the Government required the project to be delivered at no net cost to government (p2). There is not a general principle that required privately financed projects to be delivered at no net cost to government. The need for a Government funding contribution in addition to any user charges associated with a project is determined on a case by case basis, taking into account the Budget situation and the Government’s expenditure priorities.

The Government’s approval to proceed with the Cross-City Tunnel project was not conditional on there being ‘no cost to Government’. As your report points out, the Treasurer wrote to the Minister for Roads on 14 March 2002 stating that where additional funds were needed for the project they should come from the RTA’s existing forward capital program.

Secondly, your report infers that the CrossCity Motorway Consortium (CCM) won the tender on the basis of contributing the highest up-front payment (p26). Whilst the up-front was one of the main financial points of comparison in the competition, there were other important non-financial evaluation criteria. CCM’s winning bid was assessed to be a technically superior bid because there would be reduced traffic disruption during construction and the design enabled an increase in the speed limit to 80 kilometres per hour.

Thirdly, your report states that the high patronage estimates of CCM’s winning bid should have received greater scrutiny (p29). The patronage estimates of all bids were interrogated in detail. In addition, the financial models of all bidders were stress tested to determine the impact of much lower than forecast revenue projections and CCM’s bid performed well on these tests.

The traffic forecasts of CCM’s winning bid were only about 6 per cent higher in 2006 than the Public Sector Comparator (a tunnel with a 60 kilometre per hour speed limit) and 9 per cent higher in 2016. Given the challenges in accurately forecasting toll-road patronage particularly in the early ramp-up stages, it would have been difficult to conclude that these estimates were unreasonable. These challenges are evident from a Standard and Poor’s publication which found that traffic forecasts for various toll-roads around the world were underestimated on average by around 30% during the ramp-up phase (refer Global Project Financing Yearbook, October 2005). Whilst traffic flows are difficult to predict in the ramp-up phase, traffic on most of Sydney’s toll-roads have, after the first couple of years, been consistent with or exceeded initial forecasts.

It is also interesting to note that in contrast to the Cross-City Tunnel experience, Transurban, part owner of the Westlink M7, have made statements that their traffic forecasts “have proved extremely accurate” (refer Transurban Investor Briefing, February 2006).

Finally, I would like to thank you and your staff for the co-operative approach taken during the course of the audit and the opportunity to comment on the report.

(signed)

J Pierce

Secretary

Dated: 23 May 2006

Response from the Department of Planning

I refer to your letter of 5 May 2006 and the attached “Performance Audit – The Cross City Tunnel Project”.

You would be aware that the Department held discussions with the Audit Office at the officer (2 May 2006) and executive level (4 May 2006) to outline the issues raised in its review of the performance audit.  These issues were outlined in a letter dated 5 May 2006.

The Department has reviewed the “Auditor-General’s Report Performance Audit – The Cross City Tunnel Project – Final Report to Agencies” dated 5 May 2006.  The Department’s key issues are outlined in Attachment A.

I thank you for the opportunity of commenting on the Performance Audit report and trust that the Department’s comments are of assistance.

(signed)

Sam Haddad

Director General

Dated: 24 May 2006