Distribution of population growth within Victoria, 2001-2011 (Source: Plan Melbourne)

Regional development is one of those grand policy ideas that never seem to happen and yet never go away. Having recently been dispatched as irrelevant to the case for east coast high speed rail, it came back last week as one of the headline features of the Victorian Government’s new metropolitan strategy, Plan Melbourne.

It’s a politically seductive idea but as I’ve explained before (see here and here), regional development has a fraught and sorry history in Australia. However rather than seek to bring basic industries to regional centres, Plan Melbourne proposes the far simpler strategy of attracting some of Melbourne’s commuters to the country.

Under the banner A State of Cities, it proposes that regional centres within commuting distance of Melbourne will absorb some of the expected 2.5 million increase in the capital’s population between now and 2050.

Regional centres that are well connected and within viable commuting distance of capital cities offer attractive housing and employment opportunities, and so relieve pressures on capital cities by absorbing some of their growing population.

Over the period 2004-2012, around 40% of all new dwellings constructed in Melbourne were located in the Growth Areas. A further 18% were built in established outer suburbs, so it seems like there’s plenty of pressure to be alleviated.

Although it might dismay “tree-changers”, giving regional cities a bigger share of the State’s population growth (Melbourne gets 88%) would also boost regional economies (1).

This is a plausible spatial pattern. London, for example, has around 40 ‘overspill’ or satellite towns like Hemel Hempstead and Milton Keynes lying beyond the green belt around Greater London. They’re closely tied economically to the centre and are essentially London’s outer suburbs (2).

A State of Cities is an interesting idea and seems to make sense given that some home buyers are already leaping beyond Melbourne’s boundary in search of cheaper housing. But I’m not sure the advantages are so obvious.

If the policy were to be pursued on a scale sufficient to give serious “relief” to Melbourne, it would require a big investment in multiple high-capacity inter-regional road and rail links.

But all it would effectively do is transfer suburban sprawl from the outskirts of Melbourne to the outskirts of regional centres.

Indeed, because small towns and cities offer greater local accessibility, there’s a good chance residential development in regional centres would be at lower average density than the 18 dwellings per Ha Plan Melbourne envisages in Melbourne’s growth areas.

Commuting times to Melbourne would be considerably longer and more resource intensive than they would be from Melbourne’s fringe suburbs. CBD workers would use rail but since more than 70% of Melbourne’s jobs are over 5 km from the city centre, commutes by car would also be appreciably longer. Social trips to Melbourne would also involve long car trips.

Then there’s the loss of economies of scale in providing major services like hospitals, universities and specialised schools. Most new settlers will either have a lower level of service than they would otherwise have had in Melbourne or the Government will have to spend more to provide specialised services in more locations.

The impact on agriculture and rural landscapes could also be worse. Most farming land on Melbourne’s immediate periphery, particularly in the north and west (likely to take most future growth), has been degraded by proximity to urban uses and is mostly used for low value activities. Land around regional centres is likely to be less degraded.

Plan Melbourne claims that “the cost of servicing residents in regional centres is in many cases significantly below the cost of servicing residents in the growth areas of capital cities”. I think that’s doubtful and I’d like to see evidence to support the assertion.

It might be true in the very short run in places where there’s currently so-called spare infrastructure capacity but that would soon run out and in any event it isn’t “free”. Moreover many Victorian regional centres have much more serious water supply limitations than Melbourne.

The cost also needs to be compared against other options for managing growth like actively increasing the proportion of new dwelling construction absorbed by established suburbs

It’s not necessarily all downside though. Faster transport connections to regional centres would provide more housing choices for Melburnians. Creating regional commuter dormitories would bring benefits to many regional residents.

All those new citizens would require local doctors, teachers and tradies (just as they would if they’d settled in Melbourne’s outer suburbs). That’d increase the range of jobs and give younger people more reasons to stay.

Curiously, Plan Melbourne doesn’t say how much of the expected population growth to 2050 would be absorbed by regional centres. While it recognises the need for better inter-regional transport connections, it doesn’t nominate any specific infrastructure improvements either (as it does for Melbourne).

Although Plan Melbourne took over two years to prepare, I don’t think the State of Cities proposition is backed-up by any serious analysis of the costs and benefits of the policy.

While there might be other justifications for regional development, policy-makers concerned with managing capital city growth should ensure they know if the net cost of channeling growth to the regions is likely to be lower or higher than the best alternative.


  1. Prospective regional locations cited in Plan Melbourne include Ballan, Bacchus Marsh, Kilmore, Broadford, Warragul-Drouin, Wonthaggi, Geelong, Ballarat, Bendigo, Shepparton and Benalla. Restricting overspill development to suitable centres might be difficult (e.g. see here).
  2. This is why claims that cities like London can grow significantly without expanding their boundaries should be read with care.