Population projections for capital cities, high and low scenarios (source: ABS data)

By 2051, Sydney’s population could reach as high as 7.95 million and Melbourne’s 8.56 million according to projections by the ABS. Brisbane could be as big as Sydney is now by then and Perth could be larger than Brisbane.

A key task of state governments is to prepare strategic plans showing how their major cities would accommodate growth on this sort of scale. The latest document is Plan Melbourne, released in its final form by the Victorian Government last month following three years of preparation and consultation.

Plan Melbourne’s key purpose is to show how a near doubling of population could be accommodated while maintaining the city’s liveability. It assumes a ‘moderate’ scenario where population almost doubles to 7.8 million residents by 2051.

Like most metropolitan strategic plans prepared over the last decade for Australian cities, it envisages the established suburbs will have to accommodate the lion’s share of new dwelling construction. Most new housing will consequently be town houses and apartments.

It promises that transport infrastructure will be improved to deal with the higher densities that redevelopment inevitably implies.

There’re differences in detail and emphasis, but overall Plan Melbourne is pretty similar to the strategic plans prepared under the former Labor Government, Melbourne 2030 and the subsequent update, Melbourne @ 5 million (“@” cringe!).

The wonderful thing about strategic plans from a politician’s point of view is they’re mostly about the long term. So bones can be thrown to practically every lobby group without the difficulty of actually having to fund them or deal with the political fallout that inevitably comes with implementation.

There’s way too much to say about Plan Melbourne in a single article, so this time I’m just looking at two related high-level issues.

The first point I want to make is that the population projections shouldn’t be taken literally. They aren’t forecasts; they’re merely extrapolations of current trends. As the preamble to the technical document describing the projections points out, they aren’t predictions of the future.

They are not targets, nor do they reflect the expected effects of current or future policies.

The projections are based on explicit assumptions about rates of births and deaths, interstate migration, and overseas migration. The latter is by far the largest component, but it can vary over time due to a range of difficult-to-foresee factors e.g. changes in government policy, changes in domestic (pull) and overseas (push) economic conditions, changes in the attractiveness of competing locations.

Population projections for Australian cities have a history of bouncing around. For example, back in 2003, Melbourne 2030 assumed a Statistical Division population of around 4.4 million in 2031. Eleven years later, Plan Melbourne assumes the population will be much higher in 2031; around 5.8 million.

The timing of growth is very important, but the best we can do is focus on how the city could deal with a much bigger population whenever it happens. Sydney and Melbourne might reach 8 million by 2051 or they might not; indeed, that size might be so far in the future that it has little relevance for current purposes.

The other point I want to make is that apart from the obligatory one line nod to the usual suspects like the ageing population and the Asian Century, Plan Melbourne is essentially silent on the sort of future it’s ostensibly planning for. It has virtually nothing to say about the economic, technological and social changes that might occur over the next 40 years.

Yet just considering transport, improvements in computing technology and communications could potentially have a major impact on travel, from better traffic management through to autonomous cars.

David Levinson recently canvassed a range of changes he thinks are likely to shape the way we travel in the future (see also What are the big trends shaping transport?). Tyler Cowen summarises (just) some of Professor Levinson’s points:

Just as it was once standard for U.S. workers to work a six-day week, Levinson imagines that the workweek will continue to shrink…But it’s not just the workweek that will change. The pattern of how we work over the life course will also change. Levinson envisions a world in which almost half the population doesn’t enter the paid workforce until age 30…The changing workweek causes the value of office buildings to plummet…Shopping, once a big contributor to vehicle trips, is transformed as people (and their autonomous agents) order online and have goods delivered; decentralized manufacturing and 3-D printing on-demand, in turn, shrink supply chains.

As I’ve noted before (Can outer suburbs be more adaptable for future generations?), it’s hard enough to unpick history with the benefit of hindsight; but prediction is an order of magnitude more difficult. Yet if we don’t think about the future we’re planning for we’re implicitly assuming it’ll be just like the present and the recent past.

The difficulty of prediction highlights the need to devise plans that emphasise flexibility and are designed to accommodate or respond to unforeseen change.

City policy-makers should focus on removing impediments and frictions to adaptability, such as hefty stamp duty on property transactions. Institutions should be open and accountable. Prices should be transparent; they should reflect real costs rather than implicit subsidies. The cost of negative externalities should be internalised e.g. by road pricing. Equity matters and should be addressed explicitly and visibly.

Policy-makers who can’t even imagine the possibility of different futures aren’t well placed to devise flexible plans.