Follow logic and science or the romance and promises of snake oil

Of all the benefits flowing from the Victorian government’s decision to release publicly the business case for the East West Link motorway, one of the most important is the contribution this action can potentially make to improving the public debate around major projects.

The documents released by the Premier, Daniel Andrews, show the dangers and the rank stupidity of taking an uncritical approach to multi-billion dollar infrastructure investment decisions.

This is a crucial issue for public policy. The Productivity Commission observed in the recent report of its Inquiry into Public Infrastructure (see How much does the public know about transport projects?) that “selecting the right projects is the most important aspect of achieving good outcomes” from investment in infrastructure .

There are many examples in Australia of poor project selection leading to highly inefficient outcomes. In such cases, investment in public infrastructure is a drain on the economy and tends to lower productivity and crowd out more efficient projects.

The former Victorian government promised the East West Link would deliver massive benefits for all and sundry. However the public wasn’t allowed to see the analysis purporting to support the claims because it was “commercially sensitive”.

But when the numbers were revealed to public gaze by Mr Andrews last Monday, we got a wholly different picture. It turns out that were it to proceed, the costs of the East West Link would exceed the benefits by a large margin.

This sort of disconnect between the glamorous public promise and the hard-nosed reality is no great surprise. There’s no shortage of politicians, boosters, enthusiasts and advocates who relentlessly argue that the sorts of projects they favour should be funded despite what careful and even-handed analysis might show.

Their project should proceed, they say, because it’s “visionary”; because it’ll give us “the future we want”; because it’s “nation-building”; because it’ll eliminate “congestion”; because it’ll “save the planet”; or because it fits the biases of those advocating the expenditure of what, it should be emphasised, are public funds.

Analysis that might help understand if the net benefits really are as large as the boosters conveniently take for granted, or that might identify alternative opportunities that give a higher social pay-off, is blithely dismissed as the work of boring bean counters.

The latter are the kind of faithless dullards, the argument goes, who’re so seduced by logic, evidence and reason that they can’t apprehend the glorious future that’s promised to them. They’re dismissed as drudges stuck in today’s world instead of tomorrow’s, or mired in tedious concerns about trivia like costs instead of the promise of what might be.

This is infrastructure, the argument goes, and everybody knows that if there’s one thing we don’t have enough of – that we’ve dropped the ball on – it’s infrastructure. We should think less about today and more about tomorrow; we shouldn’t be short-sighted, we should be taking the long view; we should be more imaginative and expansive in our thinking; blah blah blah.

Well, the analysis for the East West Link shows that the claimed benefits of the East West Link could be as low as just 45% of the costs. No matter how sure one might feel “in one’s waters” that this motorway is a good thing, that’s a very hard number to argue with.

The often inconvenient thing about benefit-cost analysis when it’s done well is it imposes a discipline on thinking. It demands that assumptions about costs and benefits are made explicit; it requires that they must be defined and the attempt made to actually assess how big or small they might be.

That 0.45 benefit-cost ratio includes changes in variables like travel time, road accidents, vehicle operating costs, and pollution. Depending on the sort of project, the methodology is now routinely extended to include the benefits and costs from other appropriate variables like agglomeration, exercise, mental health, passenger comfort, and more. (1)

This analysis of the annual Melbourne Formula One race (Is the Grand Prix really worth $60 million?) shows how conceptually difficult ideas (in this case the value that should be accorded to the oft-touted benefit of “overseas exposure”) can be addressed analytically. It also shows there’re traps in drawing simplistic conclusions (for example, about the “employment generation” benefits of a project) that could be avoided by more rigorous thinking.

Some of these are harder to measure than others but the process is invariably better than the all-too-common posture of implicitly assuming – based on some inner half-defined and barely scrutinised mental narrative – that the benefits exist, that they’ll be captured by the project, that they’ll exceed the costs, and that who gets the benefit and who pays the cost can be ignored.

Benefit-cost analysis has its limitations and, as the Napthine government’s handling of the East West Link shows, it can be misused. But that’s true of almost anything that politicians get their hands on.

What matters most is having a way of thinking about projects that’s based on a rigorous and consistent discipline, not on vague and invariably ignorant and/or self-serving claims. As the current case suggests, a key way to improve project evaluation and selection in a politically charged environment is to insist the process is transparent.

There’s a lot at stake with multi-billion dollar investments. The Grattan Institute makes the point (see Is investment in transport a game-changer?) that even if infrastructure is productive on average,

it is only economically productive if it is “the right infrastructure, in the right place at the time and accessible at sensible prices”. Simply spending money on infrastructure is not enough to get a return if the cost benefit case is not there.


  1. Contrary to the inferences made by The Age, there’s nothing “dubious” about including variables like agglomeration effects in benefit-cost analysis, or extending the scope of the project to include “complementary” works. The issue with the East West Link business case isn’t with these concepts but with operational issues like how they were measured and how much weight was accorded to the results.