When the Western Australian Liberal government promised to build the Forrest Highway during the state election campaign of 2001, voters were told that a $136 million investment would see the bypassing of Mandurah and the saving of 20 minutes when travelling between Perth and Bunbury.
Before building began, however, the project’s estimated cost had blown out to $337 million as of May 2005, $370 million as of August 2005, $511 million in 2006 and $631 million in 2007. During construction, the price increased further to $705 million before finally coming in at $688 million – 406 per cent above the first cost figure which taxpayers were promised.
To be sure, the actual road built was in fact 20 per cent longer than that envisioned at initial stages and government documents show that $40 million in extra money was allowed in order to fund a change in materials from those which were originally specified. Adjust for those changes and the actual costs of the initial road as per its initial design was in fact around $500 million.
That’s still, however, $364 million above the initial estimate. Reasons for the remaining blowout are not made clear by official documents. What is believed, however, is that the $136 million estimate which taxpayers were given up front at the 2001 election was nowhere near plausible.
Things were not much better in respect of the Hunter Expressway in 2012, where taxpayers were told the road would cost ‘more than $335 million’ as of 2002 after the preferred route was decided in 2001. By 2009, a review commissioned by a reluctant state Labor government revealed the cost had blown out to $1.7 billion. Eventually the project cost $1.678 billion. Whether or not the project benefits exceeded its costs is unclear as no final cost benefit analysis was ever published.
Compared to those examples, the Alstonville Bypass in Northern New South Wales was arguably a success, with a cost blowout against what was first promised equating to a mere 162 per cent. First committed to by the Federal Government in 2002, the project was included in an election promise by then NSW Premier Bob Carr in 2003 for completion at a promised cost of $36 million by a promised completion date of December 2006.
Six years passed before the project was confirmed in 2009 and a construction contract worth $101 million was awarded. After $6.7 million worth of savings were made during construction, it was declared to have come in ‘under budget’ after being completed for around $93 million in 2011, six months after the contracted completion date and five years after the initially promised completion date in 2006.
Sadly, these projects – each documented in a Grattan Institute report released last October – are not isolated cases. In the report, the Institute looked at the entire portfolio of state and federal transport infrastructure projects valued at $20 million or more which have been planned or built since 2001. When compared against costs which were first quoted to taxpayers, it found that actual development costs exceeded what was promised by a cumulative total of $28 billion, or 24 per cent over and above the cumulative amounts initially promised.
This is not the result of a series of small overruns on each project. In fact, more than half of the projects came in within budget and nine per cent came in under budget. Rather, the problem lies within some of the spectacular blowouts such as those in the examples outlined above. In fact, whilst projects which came in at 50 per cent or more over what was initially promised accounted for only 17 per cent of all projects by number, these accounted for 90 per cent of the aggregate value of the cost overruns.
According to the report, further analysis of such projects reveals an underlying phenomenon in premature announcements, which occur when projects often in the lead up to an election are announced without much homework having been done in terms of benefits and costs. Projects which were announced in this manner accounted for just 32 per cent of all projects but 74 per cent of the value of cost overruns, Grattan said.
Report author Marion Terrill, transport program director at Grattan Institute, says projects can be divided into two categories: those for which adequate background work has been done prior to the announcement and those for which such work has not been done.
“There are basically two different types of projects,” Terrill explained. “Those projects which are announced with a reasonable amount of development work behind them tend to go fine. They tend to go more or less on budget and have a smoother path.
“The other type of project is one which is announced before the work has been done. A typical scenario is that a politician would promise to build a particular highway for a particular cost and they will often make this kind of promise in an election campaign. These are a different kind of project. They are bigger to start off with – on average they are bigger projects. By the time they get to a formal budget commitment, the costs have gone up by an average of 25 per cent.
“You might think it would be all over and done with at that point and that they (the prematurely announced projects) would then be on an even keel. What I found is that they (these projects) continue to be haunted by cost escalations throughout their whole lives. So by the time construction has been completed, costs have then gone up by a further 26 per cent.
“It’s a much more fraught kind of project.”
According to Terrill, this has consequences in terms of projects being built for which the business case may have been weak. Once a project has been announced, she said political considerations often dictate that projects will proceed. This is the case even where announcements have been made prematurely and subsequent information shows that in fact the project is not viable, she said.
In addition, as there is rarely any public release of actual project costs against these initial cost estimates, politicians are rarely held accountable when this type of thing takes place, she added.
To address this, Terrill would like to see all project announcements accompanied by the simultaneous release of business cases prepared by Infrastructure Australia in the case of Commonwealth funded projects or the equivalent state based body in the event of non-Commonwealth funded projects. Where projects are announced prior to business cases being ready, IA or the equivalent state body would issue a blank template, which would serve as an explicit sign to voters that an announcement had been made without the relevant diligence having been performed. This, she said, would help to promote greater discipline in terms of announcements having been made.
To help ensure that these are prepared in a robust manner, a body such as the Productivity Commission would publish independent assessments about the quality of each business case being released.
Another change which Terrill would like to see would be a new requirement that all projects worth $1 billion or more be approved through separate legislation by parliament. This, she said, would help to prevent the worst of the political polarisation which has dogged projects such as the Perth Freight Rail Link and Melbourne’s East West Link and to encourage the building of bi-partisan support for projects up front.
This, she said, is important. Politicians are spending taxpayer money on potentially city changing projects. Asking them to obtain a level of community agreement before these developments proceed is not unreasonable, she argued.
At the back end, Terrill would also like to see the publication of post implementation reviews (which the Commonwealth requires now but does not publish) and the creation of a public databank where this information could be accessed. As well as improving accountability with regard to public money and enabling the public to see where value for money was and was not delivered, she says this would facilitate greater learning from past projects and improved cost estimation processes for similar types of projects going forward.
A slightly different perspective is offered by Nolan Bear, national chairperson of the Australian Cost Engineering Society. According to Bear, one deeper issue at play revolves around critical concepts surrounding cost engineering as part of the engineering profession.
In the past, Bear said, the predominate focus of engineers revolved around design. Not enough emphasis, he said, was placed on engineers’ role as builders with capabilities not only in design but also in managing and delivering upon projects to within time and budget constraints. As a result, the importance of having cost estimates on complex infrastructure projects being performed by specialist cost engineers who had the prerequisite skills in complex data analysis has not been afforded the recognition it deserves.
Fortunately, he says, this is now changing and professional bodies such an Engineers Australia are now making a strong push for cost engineering along with other project management disciplines such as risk engineering and project control to be afforded more of the recognition and prominence which these disciplines warrant.
In addition, Bear says it is important for politicians to avoid making too greater political commitment to projects in early stages and instead to be prepared to revaluate projects at different stages as more money is expended on refining the project scope. He says politicians need to educate the public about how responsible decision making in fact involves project viability being evaluated at numerous stages prior to final commitments.
“What governments tend to do is they get engaged with a project and then they don’t’ feel that they can back out,” he said. “Governments need to understand that it is a gate process and a research process and that proper, prudent management of public money is to go through these gates and be prepared to say ‘yes, this is now not worth doing or no, we thought it was worth doing but now it’s not.'
“They need to educate the public that this is as much a reasonable and responsible announcement and position as it is to say ‘I will do what I said I was going to do’. It should be 'I will do what I said I was going to do if it is worth doing.' That’s a qualifier which politicians need to learn to included when they put in their statements.”
Bear supports Terrill’s recommendation with regard to the publication of post-implementation evaluations of projects. He says the learnings associated with this is critical from a viewpoint of improving risk management on future projects.
“What happens with these projects is that you get to the end of it, everybody leaves and the history is often not written and not recorded and all the data disappears down the drain into records which are not available,” he said.
“We need a proper culture of historic reporting that is made public on these public projects. We need a systematic recording of the results of these jobs over the initial price, the final price, changes which contributed to the final price, the changes in schedule – all of these things go into future projects. If you can see the historic overruns and underruns, you then have a basis for defining the risk on similar projects in future.”
With positive action going forward, taxpayers will be given much more robust and realistic estimates about likely costs of infrastructure projects on an initial basis.
For now, however, it seems that too often, they are being fed unrealistic estimates by politicians who have not done their homework up-front.