The best way to fix Australia’s road and rail might be out of left-field.

Australia has a problem with Infrastructure. We keep building the wrong things.

We spend a huge amount of time developing proposals that have benefit cost ratios less than one. Then, for want of alternative proposals, we turn those proposals into reality.

There are many reasons for this – politicians serving certain electorates, powerful lobby groups, bias to action, and inability to fix infrastructure through pricing .

But part of the problem is a lack of options. We build the East-West tunnel in Melbourne because it is the only idea that’s been properly developed and discussed. We put Sydney’s new airport at Badgery’s Creek because it’s the one location that has been kicked around for years. We plan a light rail line up the middle of Canberra, because that concept has been publicly flogged since Burley Griffin.

To find one great infrastructure plan, you need to discard 99 good infrastructure plans. But Australia doesn’t have 99 to throw away.

There’s a lot of talk about developing a “pipeline” of infrastructure ideas. But politicians are very risk averse, and big infrastructure companies don’t want to waste money on business plans. So our pipeline is the diameter of a carpet python, with a couple of big lumps where it has been fed an approved mega-project.

Infrastructure Australia was set up by the Rudd government to try to help develop a pipeline of ideas, and independently test them. But it only has a few staff, and its leader was recently fired by the Abbott Government. Just as that was happening, he took a stand, publicly saying ”Entrenched truculent bureaucracies have impeded progress… It has been heard that some good ideas cannot go ahead because they would set ‘precedents’. Among other things, this implies knowledge of, but unwillingness to address, widespread deficiencies. Such wilful attitudes test the patience of our elected masters, industry, and the public.”

Even with this courageous bureaucrat in charge, the old Infrastructure Australia was unable to renew the infrastructure planning system. With his blood all over the carpet, the new Infrastructure Australia is cowed.

Even if it makes our political leaders uncomfortable, Australia desperately needs a truly independent infrastructure development and analysis capacity. But how?

I think the answer is not a bureaucracy. I think the answer is by using technology. The same technology that has helped humanity create the biggest encyclopaedia in history and a hugely detailed map of the entire world.

A Wiki.

Imagine a website where you can start a page for any infrastructure project you might dream of.

• You want to extend a train line by a few kilometres? Start a page with a description and a map.
• You want a helicopter pad installed at the local sports ground? Start a page with a project description and a map.
• You want to build a very fast train between Melbourne and Brisbane? Start a page with a project description and a map.

Each page would be able to be edited by absolutely anybody. There would be a section for environmental impacts, a section for cost estimates, a section for estimating time for planning and building, a section for land-use changes and implications, a section for creating a cost-benefit analysis.

Most pages would be the hare-brained schemes of the lone wolves of suburbia. The pages would be underdone and silly. But by asking the community to rate each page, the better ideas would attract contributions from a range of talented people and rise out of the muck.

This could bring unforeseen solutions out of obscurity.

For example, when I was writing about the ridiculousness of the state government’s plan to put a new railway station in South Melbourne, right on top of an existing tram stop and miles from the new development it ostensibly serves, I found in the depths of the internet forums the suggestion that the problem could be solved better with a train line that runs from north-east to south-west.

“Fisherman’s Bend should be on a new line from Merri (Northcote) to Newport (Wydham Vale – Mernda line).”

This idea may have a cost-benefit analysis ten times better than all the existing plans. But how would we know? There is a choke-point for shining light on new ideas, and its name is the Department of Transport. Risk-averse and slow-moving, DoT can only be expected to properly consider a few ideas that it thinks the government is interested in.

Choosing between a big group of well-developed projects that each have a range of intriguing benefits and positive cost-benefit analyses is going to be difficult, politically. But it’s a better problem than the one we have now, which is a small group of infrastructure plans that mainly look like wastes of money. (I refer here to Melbourne’s East-West tunnel and airport rail, but an honourable mention should go to the plan for a very fast train up the east coast)

If you want to find one project with a cost benefit analysis good enough to build, you need to look at 100 or 1000 projects across the country. The current pipeline is starved of proposals, so it is no wonder the infrastructure policy space is so sickly and anaemic.

PROBLEMS:

1. The majority of analyses on the wiki would be defective. But the community should privilege the ideas that have the most potential and these should attract rational people to contribute. I expect projects will be submitted with wild underestimates of their cost, and there will be a push by more rational people to actually use cost estimates that reflect realistic Australian pricing. On the ‘discussion’ pages  I imagine there would be fierce argument about why it is we can’t have projects delivered at the prices that apply in China, America, etc. Projects with decent estimated benefits should be most willing to use realistic pricing.

2. Monorails and personal rapid transit. It’s going to be hard to keep really wild ideas out of there, but perhaps that’s the point.

3. Splintering into too many pieces. Every little change in a proposed plan (should this freeway have an exit here or here?) would potentially lead to a new page being created, with a new set of costs, environmental impacts, etc, etc. Conventions and rules may need to be developed to guide when a change is big enough to warrant a whole new entry. But wikis are good at developing cultures and rules that make them effective.

4. The whole thing would be at risk of being ignored if there was no suggestion governments would at least look at it. The project would be a really great thing to seed with some official resources, for example, freely available mapping software for wiki users to use, some models for doing traffic and demand forecasting, recommended ranges for cost per kilometre of roads, bike lanes, tram lines etc. I’d like to see it hosted at infra.wiki.gov.au to give it a sense of official imprimatur and encourage involvement. Perhaps the government could loose a few bureaucrats or pay a few infrastructure experts to play with the wiki to get it started, anonymously having them make edits and bring in a bit of rigour.

So, is this a good idea or a mad one? Is there some aspect of wikis I have overlooked or some problem I’ve not foreseen? Leave a comment below or hit me up on Twitter!