Issues Magazine

Population, Planning and Perhaps Getting it Wrong

By John Minnery

An example of City of Brisbane planning in the 1940s illustrates that population projections are sometimes wildly inaccurate.

Urban planning as a profession relies heavily on estimates of future populations. Both governments and private developers may have to commit millions of dollars on the basis of these estimates.

Urban planners have to advise both governments and private developers. For example, shopping centres need to be located where they are accessible to large numbers of future shoppers; if they are built where there is expected to be a large number of people in future but the projections turn out to be wrong, the development companies may face bankruptcy.

Governments need to plan the future location of schools well before they are needed but precious financial resources are wasted if the population projections are inaccurate.

Similarly, people moving to newly developed housing areas expect water, sewerage, electricity, telephones and the like to be available when they arrive or very soon afterwards. To construct these services in a timely manner the providers have to guess where future new populations needing their services will be located because it takes a long time and a lot of money to build a new sewerage system or to put in water mains.

The impact of increasing population numbers and their infrastructure needs can be seen on the developing fringes of most Australian cities (Fig. 1).

Three aspects of future populations need to be estimated. The first is the actual number of people. This is partly what the current debate about Australia’s future population is about. How many people can the continent support?

But equally as important in planning terms is the second aspect: the make-up of that population. What proportion will be aged over 65 and likely to be retired? What proportion will be between 5 years and 18 years old and thus likely to need schools? How many of the future population will be living together in how many households, and what size will these households be (and so, how many dwellings will they need)?

The third aspect is the connection between population and other aspects of life. This involves trying to estimate things like the average number of cars used by each household (and thus the impact of the population on the need for roads), the proportion of households with low incomes who will need more affordable housing, or whether the future population will want to live in smaller houses than people live in now.

All of these aspects of projecting, or estimating, future populations are important. The trouble is that demographers and planners do not always get it right. An example from the history of Brisbane illustrates this.

The current City of Brisbane was formed when about 20 previous local governments (including a much smaller original city of Brisbane) were amalgamated in 1925. This made Brisbane quite different to the local governments in the other metropolitan areas in Australia. The new enlarged local government included all the urban land in the metropolitan area and a considerable area of surrounding farming and vacant land. A plan for Brisbane would be a plan for the whole metropolitan area, as well as a plan for the future metropolitan population.

Planning in Brisbane was a little slow to become effective. But in 1937 the city planner, Mr R.A. McInnes, was asked by the Brisbane Courier Mail what he thought Brisbane would be like in the year 1987, 50 years into the future. His predictions were published on 4 May 1937, then repeated in an article called “The dangers of guessing the future” in the Courier Mail on 16 May 1979.

Mr McInnes started by noting the huge changes that had occurred over the previous 50 years, but then showed how the area’s population had increased from 83,000 people in 1887 to 314,000 by 1937 – “nearly four times as great”. He thought, however, that this rate of increase could not continue. He noted that the birth rate was then falling or steady, even though the death rate was also falling. So he felt that unless fresh people came from other places, “there is no tangible reason for expecting a population greater than 350,000”. The population estimate was important to him, because he felt that “in trying to visualise what Brisbane will be like in 1987 this is one of the most important facts”. Brisbane’s population in the 2006 census was close to one million people, so the population estimate was not a fact but a wildly inaccurate guess.

This population estimate formed the basis for McInnes’ proposal for a “green belt” to surround the then built-up area, with satellite towns located outside the green belt but inside the city’s boundary. This form for the metropolitan area was reinforced through later inaccurate population predictions. One made by a later officer in charge of city planning, Mr F.G. Costello, in 1949 was based on an estimated ultimate population for Brisbane of 750,000 people.

The green belt plan for Brisbane was discussed by the council and the public in the 1940s and 1950s, but never actually implemented. Two versions of the plan can be seen in Figures 2 and 3, from the Courier Mail in 1946 and 1953.

Clearly the various estimates of the future population of Brisbane were wrong. Not only has the population of the city grown to almost one million, but urban development has now spilled out over the council’s boundary into the neighbouring local authorities. We now plan for the wider South-East Queensland region rather than just for the city itself. According to the Australian Bureau of Statistics in 2008, this wider region had a population of 2,657,300 in the 2006 census.

Urban planners and demographers have developed a number of ways of dealing with the uncertainties of projections of future populations. One method is to publish high, medium and low range projections based on slightly different assumptions about future growth.

When some years ago I worked as an urban planner in a somewhat unsettled region in Africa, the plans we produced for the city we were working in were based on turning the idea of projecting population on its head. Instead of trying to estimate what the population would be at different times in the future, we estimated the land uses that would be required at different increased future populations and based our planning on these. No attempt was made to link the population to particular years. In other words we said: “When the population reaches 250,000 you will need this much land for housing, this much for industry, this number of schools, and so on. We can’t tell you when that will be, but if you monitor population growth you will know when the facilities are needed.”

So projecting future populations is very important. It affects a whole range of things we need to plan for. But we need to be careful because we can sometimes get these projections wrong. Just look at the case on Brisbane in the 1940s.