Infrastructure: No longer a no-brainer

One of the popular economic memes of the 2000s has been that Australian needs more infrastructure. It has filled out many a think-tank report. In the form of the National Broadband Network, it helped Labor win government in 2007. It has led to a current crop of serious proposals for truly dumb projects, such as Melbourne’s Avalon Airport Rail Link, and merely dubious ones, such as Melbourne’s East-West Link freeway.

We have also built a great deal of infrastructure in recent years. And while we still have plenty to build, it may no longer be true that we have a generalised infrastructure shortage.

Some of the signs are captured in the graph below. It comes from a McKinsey Global Institute (MGI) report called Game changers: Five opportunities for US growth and renewal. MGI is the McKinsey the management consultancy’s macroeconomic think-tank. Be warned: it is not always the most rigorous of outfits, and its sources and methods for this report are not clear on a first reading.

From McKinsey & Co's "Game changers: Five opportunities for US growth and renewal"

From McKinsey & Co’s “Game changers: Five opportunities for US growth and renewal”

For what it is worth, MGI is making the point that the US could use a deal more infrastructure. But tangentially, over on the right-hand side of the graph, MGI also makes the case that Australia is not in the same boat as the US. The graph suggests we are actually spending more than we need to do – just like that famously profligate infrastructure spender, Japan.

At the very least, if MGI’s figures are broadly sensible, we are not obviously underspending on infrastructure in general.

The graph tends to support the claims made by the Grattan Institute’s John Daley last year that “investing faster in Australian infrastructure is unlikely to substantially increase the size of the economy over the next decade” and that “Australian infrastructure spending is already at historic highs”. Daley, an ex-McKinsey executive, made the claims in a report called Game-changers: Economic reform priorities for Australia, which Club Troppo’s judging panel – me – found deserving of an Ozzie Award.

As the Grattan Institute report pointed out, overspending on something like infrastructure is far from costless. It leaves less money for everything else. (In Australia’s case, a quick look at the MGI report suggests the something on which we are underspending might be education.)

Why have we had all the scare stories about Australian infrastructure? They typically start with the Engineers Australia Infrastructure Report Card, where the nation generally gets a C mark and a mournful complaint that “little or no real overall progress has been made”. (Engineers Australia is the local engineers’ union/lobby group.) Hired economists then take this supposed infrastructure deficit and put some very loose, but large, dollar numbers around it. It has never been a very rigorous process. It has, however, attracted plenty of money over the years, some of it from large civil engineering firms.

Labor has not covered itself with glory on infrastructure policy. The most obvious example is the NBN, which badly needed some sort of independent assessment – an assessment which would likely have revealed it to be an inefficient way of improving broadband services. But by introducing Infrastructure Australia as an independent assessment body, Labor has at least created an institution which can rein in over-eager infrastructure spending. On that, its timing seems to have been exactly right.

Nothing in the above assessment means we should stop spending on infrastructure or working to make infrastructure planning and construction better. There’s a lot to be done, especially in places like south-east Queensland. This assessment does, however, suggest that we should pay attention to Infrastructure Australia’s processes, and make sure to assess infrastructure proposals with a critical eye.

About David Walker

David Walker runs editorial consultancy Shorewalker DMS (, editing and advising business and government on reports and other editorial content. Newsletter: . Among other roles, David has edited the award-winning Acuity and INTHEBLACK magazines, been chief operating officer of online publisher WorkDay Media, held senior policy and communications roles at the Committee for Economic Development of Australia and the Business Council of Australia and run the website for online finance start-up eChoice. He is a former economics writer for The Age and News Ltd. He has qualifications in law and corporate finance.
This entry was posted in Economics and public policy, Interesting Graphs. Bookmark the permalink.
Notify of

Newest Most Voted
Inline Feedbacks
View all comments
Bruce Bradbury
Bruce Bradbury
10 years ago

Does ‘infrastructure’ in their report include private infrastructure spending? More specifically, is the high spend for Australia due to mining infrastructure spending?

10 years ago

Engineers Australia is neither a union, nor primarily a lobby group.

Perhaps you might like to address the substance of the report card, which is an attempt to identify where there are infrastructure problems and prioritise them.

For example, you may not agree with the C ranking for whatever reason, but at least you have information broken down by State and Territory, showing where standards of infrastructure are relative to each other. So you then might say why you think a particular rating in an area that you are familiar with is wrong.

I would point out that in the NT, this report card did point out the need for investment in power infrastructure…and was ignored by people who thought that Engineers Australia was a union/lobby group. That was until there was a catastrophic failure in the system a few years ago leaving large parts of the metro area out of power for a long time. Of course, since people’s memories are short, now that Power and Water in the NT has invested that money to address the infrastructure shortages, they have been given a bollocking for over spending by the accounting types. Should I say “Short term, short sighted accountants…sigh” or would that be impolitick? ;)

FWIW, APESMA is the union for engineers.

Robert Merkel
10 years ago

I think you hit the nail on the head in this sentence:

Be warned: it is not always the most rigorous of outfits, and its sources and methods for this report are not clear on a first reading.

Maybe Australia is overspending on infrastructure. Maybe it’s underspending. Either way, I don’t think you can draw any useful information from that graph.

I wonder whether they took into account what is widely perceived as an infrastructure spending drought through the 1990s by state governments obsessed with AAA credit ratings.

derrida derider
derrida derider
10 years ago
Reply to  Robert Merkel

Yep, Oz may or may not be spending too much on infrastructure – but this chart is not any evidence of that at all. No documentation at all as to how “need” is derived here, for a start.

Data without context can never be made into evidence, as any experienced analyst knows.

Luke Elford
Luke Elford
10 years ago

It seems rather odd to present as evidence in support of one’s position the work of a think tank one doesn’t think is very rigorous, especially if the methods used in that work aren’t clear. But anyway, an indication of the method used to determine “need” is contained in another McKinsey report, “Infrastructure productivity: How to save $1 trillion a year”, which contains a graph, Exhibit 4, displaying the same results for the United States and Japan. The other countries in the graph in the post, including Australia, aren’t shown in this graph.

As detailed in the infrastructure productivity report, the method used to determine need is as follows. They used the perpetual inventory method to estimate the value of the infrastructure stock relative to GDP for 12 countries for which they had enough data. This didn’t include Australia: they present no data showing the relative size of Australia’s infrastructure stock. Discarding the results from Brazil and Japan, they averaged the values for the other ten countries. They took this average, 70%, and assumed that it applied to the world as a whole. They then calculated what percentage of world GDP would need to be spent on infrastructure over the period 2013-2030 to maintain the 70% result, given forecast GDP growth and an assumed depreciation rate of 2.5%. They then appear to apply the same method to individual countries.

In other words, they appear to assume that Australia’s infrastructure stock is 70% of GDP, assume that maintaining this ratio is optimal, and calculate required investment spending on this basis. This, then, is an exercise in mathematics, not economics; there is no grounding in benefits and costs.

10 years ago
Reply to  Luke Elford

Quite right.

In addition, it takes no account whatever of the efficiency or effectiveness of that infrastructure.

Luke Elford
Luke Elford
10 years ago

A few more points:

1) The graph does not suggest that “we are actually spending more than we need to”. To accept the method they use as valid is to accept that Australia needed to spend more between 1992 and 2011 to bring the infrastructure stock up to the desired 70%.

Nowhere in either of the reports is there a reference to Australian infrastructure spending being profligate. There are plenty of mentions of Australia, but they’re positive, including this comment about the NBN: “Infrastructure planning should be rooted in broader socioeconomic objectives set through a political process, and selected projects should address those objectives directly…Australia set an objective of delivering high-speed Internet connectivity to 93 percent of its population and identified the establishment of a national broadband network as the best way to achieve that goal”.

2) It is wrong to infer something about government infrastructure spending from data on overall infrastructure spending. The evidence suggests that Australia’s ratio of government capital stock to GDP is below the average for the OECD. Kamps (2006) found that it fell from 93% of the OECD average in 1980 to 78% in 2000 ( According to the OECD, public investment in Australia averaged 2.91% of GDP per year for 2001-2006 and 3.55% for 2006-2011, compared with OECD averages of 2.99% and 3.10% (

The 2012 BITRE infrastructure statistics yearbook shows that increased infrastructure construction as a percentage of GDP since the 1980s is accounted for solely by private sector construction for private sector project owners. From the beginning of the 1990s, governments have essentially lost control of overall infrastructure spending as the relevant industries have been privatised and government agencies corporatized. If one thinks that infrastructure spending in Australia is too high, the obvious question is what went wrong in this process. There have certainly been serious questions about investment in electricity transmission and distribution, not to mention a series of failed private road tunnel projects.

3) It seems odd to argue that there is a need for Infrastructure Australia to “rein in over-eager infrastructure spending” when the project that Infrastructure Australia has identified as the nation’s top priority can’t get funding.

David Walker
David Walker
10 years ago

The participants in this thread are making lots of valuable points. I’d agree with the McKinsey analysis that compared to many other countries, most Australian investment decisions have been pretty sensible. And the correct way to assess infrastructure projects remains benefit/cost studies on those specific infrastructure projects. No macro analysis can replace that.

If you think it’s odd that I’m pointing to the limitations of my sources, that’s a sad commentary on how sources are mostly used in public debate. Part of the spirit of Club Troppo is to point out the weaknesses on both sides of an argument.

I’m not really want to dig in too hard on one side of the macro argument or the other, because part of my view is that macro analyses of infrastructure need aren’t that useful. What I am doing is pointing out that what macro evidence there is now cuts two ways, and that if you care for macro numbers, some of them say there may be less of a stock of high-benefit infrastructure projects than was the case a decade ago.

So to the extent that people have assumed Australia still has an “infrastructure backlog”, I’m saying “don’t assume that’s still true”. Like the title says, “more infrastructure” is no longer a no-brainer decision”.