The big picture on population growth.I realise I have already...

  1. 2,172 Posts.
    The big picture on population growth.

    I realise I have already posted the article on another thread, but I do feel its also relevent to the ensuing discussion on this particular thread.

    Cheers


    Population boom forces rethink by urban planners
    DEMOGRAPHER: Bernard Salt March 05, 2009
    Article from: The Australian

    THERE has been a fundamental change in thinking by demographers about the medium-term outlook for population growth in Australia.

    In September, the Australian Bureau of Statistics published a medium projection that showed the nation adding 6.9 million residents over the 20 years to 2026.

    Just two years earlier, the official outlook for this period was net growth of 3.7 million.

    The extra 3.2 million residents materialised out of new trends and data flowing from the 2006 census and changes to migration policy that upped the net migration assumption.

    The bottom line is that the birth rate was lifted and the assumption regarding overseas migration was also lifted from 110,000 to 180,000 per year.

    What this effectively meant was that strategic plans that had been developed for capital cities between 2002 and 2005 (and based on 2001 census results), including documents like Melbourne 2030 and Sydney's City of Cities, had to be rethought.

    And that is precisely what's happening right now.

    New plans are being developed for raised expectations of growth.

    But the problem has always been that the ABS develops this national view about population growth and it is then left up to the states to work out how and where growth is to be accommodated.

    And the states are doing exactly that: new population forecasts are being prepared which allocate growth, based on preferred planning principles, at the local government area level.

    This is an important process for the property development industry to understand because it provides an "official outlook" for the demand for housing and associated infrastructure. But the problem with the state projections is that when aggregated they didn't always tally with the ABS view of national growth. Or this was the problem until now.

    The ABS was commissioned last year by the Department of Health and Ageing to develop population projections for every local government area on the Australian continent.

    The output is the first publicly available population projection at the small area level where all the bits add to an agreed national outlook. (See www.health.gov.au)

    This is a very powerful database for two reasons.

    First, it shows the scale and location of growth across the nation and second, it carries the gravitas of a nationally co-ordinated perspective produced by this nation's foremost demographer, the ABS.

    In other words, it's a good document to cite in an argument about rezoning, or in the development approval process. But the advantage of the DoHA database goes further than working out where demand for property will be greatest: it can also indicate where current zoning must be rethought in the short term.

    Maybe you have developable land located just beyond an "urban growth boundary" or similar; this document and its logic might just be the leverage needed to change local planners' thinking.

    Here are just a few examples of where planners, working off dated projections anchored to the 2001 census, must expand their plans for housing and other infrastructure.

    The previous outlook for Melbourne showed the city adding 654,000 residents over the 20 years to 2026; the new projections lifts this number to 1.3 million. Sydney was to add 754,000 in this time frame, whereas the number now is 1.1 million.

    Strategic planning for the Gold Coast was based on the expectation of 240,000 residents being added over the 20 years to 2026, whereas the new outlook is for 365,000.

    Even the outlook for places like Wagga has been upped from 6000 under the old state projections to 16,000 under the DoHA projections.

    A net extra 10,000 residents for Wagga means a new suburb must be delivered by 2026.

    One of the issues with the DoHA database is that the projections are published by small area. We have aggregated all of the required small areas to align with the current definition of the 70 largest cities and towns in Australia. This is the first time anything like this has been done and it provides a unique opportunity to identify and prioritise property and infrastructure development opportunities.

    The town that is expected to support the greatest percentage increase in population over the projection period is Hervey Bay. This community is expected to grow from 51,000 residents in 2006 to 102,000 in 2026, which is a 99 per cent increase.

    The projections show that Hervey Bay will double within 20 years. It is already this nation's 29th largest city; by 2026 "the Bay" will rank 24th.

    Other rocket towns include Bunbury, projected to grow by 74 per cent over the 20 years to 2026, followed by Gladstone (up 68 per cent), the Gold Coast (up 65 per cent), Mandurah (up 64 per cent), Busselton (up 62 per cent) and Cairns and the Sunshine Coast (both up 61 per cent).

    The common denominator of these places is their sea-change lifestyle, although Bunbury and Gladstone have local manufacturing and port-handling capacities.

    But the projections aren't all about rapid growth.

    Some towns are expected to grow slowly over 20 years, with population increases under 10 per cent, such as Burnie-Devonport, Orange, Griffith, Grafton, Whyalla, Armidale and Innisfail.

    Towns expected to actually lose population include Goulburn, Lithgow, Broken Hill and Ayr.

    A key assumption driving the projections is that recent migration trends will be carried forward.

    Migration from small rural communities prompted by drought early this decade is implicit in many of the projections. It may well be that the fortunes of towns like Orange, for example, reverse as the drought recedes. But places like Orange might well be on the frontline of Australian demographic change brought about by effects of climate change.

    The projections contain a number of surprises.

    Victoria's Latrobe Valley was projected by old state-based forecasts to remain more or less static in terms of population levels over 20 years to 2026.

    However, the DoHA projections show the Latrobe Valley population rising from 76,000 in 2006 to 88,000 in 2026. It would appear that Gippsland planners must accommodate the unexpected inflow of 12,000 residents, although the reality is that local planners will most likely await publication of state forecasts before carving out new subdivisions.

    Dated projections for Canberra-Queanbeyan assumed that 46,000 new residents would be added by 2026, whereas the new projections lift this outlook to 107,000. Canberra planners must now accommodate an extra 60,000 residents within 20 years.

    One concern business will have about this new perspective on the small area population outlook for Australia is how the assumptions might be affected by the global financial crisis.

    For example, there is already talk of reductions in the migrant intake.

    However, the DoHA outlook assumes net migration of 180,000. In the year to September last year, this figure actually stood at 213,000.

    The fact is that net migration to Australia can be reduced by about 15 per cent before breaching the long-term assumption driving the projection. And there are important reasons why net migration should be maintained over the longer term, not the least of which is the need to sustain workforce and tax growth as baby boomers retire next decade.

    Finally, if the recession is even worse in New Zealand than in Australia, we can expect the "Tasman Tide" to again begin ebbing in our direction.

    I think the more pressing impact of the global financial crisis on the DoHA projections will be the advent of unique economic and lifestyle factors.

    The collapse of coal exports, for example, may have a localised impact on job and population growth in Gladstone, Mackay and Bunbury.

    Spectacular recovery from drought may change the outlook for Orange and Goulburn.

    Concern about the effects of UV rays may prompt accelerated migration to Tasmania.

    A prolonged recession might diminish the capacity of retirees and lifestylers to support classic sea-change communities such as the Gold Coast.

    On the other hand, Australians have shown a clear predilection for using trigger events, such as job loss, as a reason to reinvent themselves in a new community on the coast.

    Taking all this into account, my overall view is that even if migration is damped down during the recession, it must at a later date spring back to prop up the worker and tax base.

    This logic at the national level supports the aggregate of the DoHA projections over 20 years, even though there may be yearly variances.

    Nevertheless, you must decide how the global financial crisis might or might not distort the base picture presented in this newand most important set of projections.

    Bernard Salt is a KPMG partner ([email protected]; www.bernardsalt.com.au), with data analysis from James Paul, KPMG manager (jamespaul@ kpmg.com.au)



 
arrow-down-2 Created with Sketch. arrow-down-2 Created with Sketch.