Published by the Australian Financial Review, Monday 24 October

It is one of the most predictable election pictures we know: a campaigning politician in fluoro vest and hard hat, announcing a new road, bridge or rail line. But this image should do more than make us feel resigned and cynical. It exposes a vital policy issue.

For the first time, the Grattan Institute has analysed spending on all 836 transport infrastructure projects valued at $20 million or more and planned or built since 2001. Our new report, Cost overruns in transport infrastructure, shows that these projects have cost $28 billion more than politicians promised they would – about a quarter of the projects’ total value.

With the economy sluggish, private investment declining and interest rates at historic lows, many politicians and commentators have called for new infrastructure spending to boost productivity. Yet infrastructure is only worth building if the benefits outweigh the costs. Often the distortion in the claims made for Australian infrastructure projects and what they finally cost is so great that governments might have done better to have spent the money on something else.

Premature announcements are the biggest cause of cost blowouts. When ministers and oppositions announce a new highway or metro line before a formal funding commitment, these early cost estimates often turn out to be spectacularly wrong. Western Australia’s Forrest Highway between Perth and Bunbury cost nearly five times, and New South Wales’ Hunter Expressway cost nearly four times, the amount initially promised.

Analysing cost overruns from the first funding promise is not common practice, but it should be. Once politicians have announced a project, they and the public treat that announcement as a commitment, and they are right to do so: two thirds of these projects end up being built.

Large iconic projects are harder to deliver to budget because they are more complex than smaller projects, and over-optimistic cost estimates announced as election promises are likely to be flawed. Projects announced close to an election have cost overruns 23 per cent larger on average than otherwise similar projects.

All major political parties claim to grasp the problem. The Coalition, Labor and the Greens have made hand-on-heart commitments to project selection based on sound analysis and independent evaluation, to taking advice from the independent statutory body, Infrastructure Australia, and to investing in the long-term public interest.

Unrealistic promises

But their claims are hollow. All three parties at the July federal election mostly promised projects that Infrastructure Australia had not yet evaluated, or had found to be not worth doing.

Such recklessness is not tolerated in other sectors with similar levels of public investment. The $123 billion Future Fund has an independent board of guardians; its legislation requires investment decisions to be made at arm’s length from government; and it tables its annual report and financial statements in Parliament. Transport infrastructure investment by Australian governments is about the same size – $141 billion since 2000 – and should attract the same public scrutiny. It does not.

Payments to unemployed people are worth somewhat less, at $108 billion since 2000. Yet social security law lays out in exhaustive detail the conditions under which an unemployed person can qualify for payment, the rate they are paid, and the arrangements for recovering incorrect payments. Politicians frequently bemoan waste in the welfare system and the need to reduce fraud, improve compliance and get better value for money. They rarely do the same for transport infrastructure.

Solutions exist. Transport infrastructure spending should not be a slush fund. Instead, we need to curb governments from committing to projects until they have been properly evaluated and the business case tabled in parliament. We need Infrastructure Australia to report on all infrastructure projects funded by the Commonwealth, not just those that governments choose to ask them about.

The post-completion report on how a project performed against its costs and benefits should be published to data.gov.au for all to see. Finally, if a government is prepared to put up $1 billion or more for a big project, it should introduce standalone legislation to allow scrutiny and encourage bipartisanship where it really counts.

Yet the political will to make such changes seems to be missing. Even today, multibillion-dollar projects such as Melbourne’s Western Distributor, Sydney’s WestConnex and the Inland Rail between Melbourne and Brisbane have much more optimistic cost estimate profiles than history would lead us to expect.

New Reserve Bank Governor Philip Lowe is one who believes not only that our cities need new transport infrastructure but that it can be built efficiently. But, he stressed recently to a parliamentary committee, projects must “pass the strong business case and have good governance around them”. With proper project selection and governance, infrastructure can be a sharp and effective economic tool. But right now, that goal is a long way down the road.