SSEC logo Sutherland Shire Environment Centre  

Scenarios for the Future Population of Sydney

Prof. Peter McDonald Prof. Peter McDonald - Head, Demography program, Research School of Social Science, Australian National University

Most of the work that I have been doing has been at the national level, and I'm about to embark upon quite major work at a regional level, working for the Australian Housing andUrban Research Institute. They are going to be doing projections of population and housing and households at a regional level, over about seventy regions in Australia. And so I was grateful for the opportunity to start having a look at Sydney.

As that quotation from Mr Ruddock said, the situation of the cities, and Sydney, is different to the national level. And that needs to be taken into account. I'm going roughly to be telling the story that everybody is speaking about - one million in twenty years and then another million in the thirty years after that.
I'm going to go through the demographic trends first.

Demographic Trends


At 1.73 births per woman, Sydney's fertility rate is relatively high. After Darwin, it has the highest fertility rate of any Australian capital city. Also, as the chart shows, Sydney's fertility rate has levelled off over the past few years.
Births and Deaths, 1999-2049
This is in sharp contrast to Melbourne where the fertility rate has shrunk to 1.58 and is still falling. Although I have not made precise calculations, the difference between the two cities is almost certainly related to the differing ethnic compositions of the two cities.

In Sydney, the high fertility rate of recent first generation immigrants from the Middle East, the Pacific and to a lesser extent, Vietnam, contrasts with the very low fertility rates of the second generation of Southern European origin in Melbourne. You expect to get relatively high fertility rates in fringe LGAs. People live there because of the nature of the housing and because they have children. This is the case in Sydney but, in Sydney, the highest fertility rates are in Auburn, Liverpool, Blacktown, Fairfield and Canterbury-Bankstown. Ethnic composition is clearly playing a part. In projecting Sydney's fertility rate, we could expect a return to falling fertility as this high fertility first generation immigrant group shifts to older ages unless they are replaced by others with high fertility rates.So, the range of options for the future fertility (TFR) is probably between about 1.5 and 1.7 births per woman.

International Migration Trends

Sydney constitutes about 1/5 of Australia's population but about 2/5 of net overseas permanent and long-term migration goes to Sydney. Probably more move to Sydney relatively soon after arrival. For example, anecdotal evidence suggests that the refugees from Iraq and Afghanistan, once released from the detention camps, find their way to Sydney. Hence, any change in the number of overseas migrants coming to Australia has a much bigger impact on Sydney than it does on Australia as a whole.

When net migration for Australia is 90,000 (about the present level), net overseas migration to Sydney is about 34,000. If net overseas migration were to jump to 110,000 per year, the Sydney component would be about 42,000. As the main debate at present at the national level is between the present level of migration and a somewhat higher level, for the projections, I consider overseas migration to Sydney ranging between 34,000 per annum and 42,000 per annum. This corresponds with the most recent Australian Bureau of Statistics (ABS) projections.

Recent trends have been consistent with the lower end of this range.
Net Migration, Sydney, 1988-1998
While there are those who call for the diversion of international migrants away from Sydney, past schemes to attract international migrants to other places have been less than successful. Furthermore, there has been a shift in the migration stream to long-term temporary migrants and away from permanent migrants. The skilled temporary stream are highly likely to move to Sydney because of its status as an international financial centre. Thus, there seems little likelihood that the attraction of Sydney for international migrants is likely to abate.

Internal Migration

A 1994 Discussion paper entitled, Sydney's Future states that, as more people come from overseas to Sydney, more Sydney residents leave for other places in NSW or Australia. This is a view held by others as well who state that rising international migration is not a problem for Sydney because an equal number of Sydney residents would leave. The mechanism for this is said to be housing prices. The theory is that a larger international migration intake would force up the price of housing in Sydney. As this happens, young people leave Sydney because they cannot afford its rents or those who own houses sell up and move elsewhere realising a capital gain. While we might question the desirably of spiralling housing prices as a 'good thing', examination of trends in the past provides some credence to this theory.

However, there is another interpretation. The peak years of movement out of Sydney in the past 30 years have been 1974, 1981 and 1990. These are all recession years. Thus, an equally plausible theory is that movements in and out of Sydney are a function of the business cycle. As international migration drops off in recession years and internal migration increases, this gives the surface impression that internal migration out of Sydney has risen in relation to previous high international migration. The slide shows that the rise in international migration to Sydney since 1993 has not been matched by a rise in internal migration out of Sydney. Thus, I believe that we need to be sceptical of the proposition that a rise in international migration to Sydney would automatically lead to a rise in internal migration out of Sydney. Indeed, since 1993, net overseas migration to Melbourne has been rising steadily while at the same time internal migration has turned from negative 19,000 per annum to positive 4,000. There is an argument that the change in the nature of the economy and the labour force is drawing people towards the two major cities. An American researcher, Saskia Sassen, has shown that with the growth of the information economy, population has grown in the major financial centres that are the capitals of the information economy. This is counter to the notion that people can be located anywhere in an information economy.

Thus, in the projections, I have taken two possibilities for migration. The first, a gain of 22,000 persons per annum, is a continuation of present levels of both internal and international migration. The second assumption, 34,000 per annum, is based on the number of international migrants that would come to Sydney if overseas net migration to Australia rose to 110,000 and internal out migration from Sydney fell from 12,000 per year to 8,000 per year. Note, the present level of net migration to Melbourne is about 30,000 per year.

Age structures of migrants are applied separately for internal and external migration based on recent experience for Sydney.

Not in the short term but in the longer term, two factors could move net migration to Sydney out of the 22-34,000 range. First, as the baby-boom generation reaches retirement, some may take the coast option. More on this later. Second, flows into Sydney from both overseas and other parts of Australia are contingent upon Sydney maintaining its position as a world financial capital. If Sydney should lose this status, then, not just Sydney but Australia as a whole will slip into the situation now faced by Tasmania and New Zealand.


All the projections assume the same future mortality as is assumed for Australia as a whole by the ABS. In fact, expectations of life for Sydney are slightly higher for Sydney than for all of Australia at present, but the difference is not significant.


The first slide contrasts two projections.

Both set the fertility rate at 1.5 by the end of this decade. They have two different migration assumptions.

As I project the populations forward (below), you will notice that the age structure gets older, but the age structure of the two projections is very similar.

After 50 years, (below) the projection with an additional 12,000 net migration adds about 700,000 extra people to Sydney, but makes almost no difference to its age structure. At the end of the period, Sydney's population is still growing according to both projections - by about 14,000 per year in the 22,000 migration projection and by about 28,000 per year in the 34,000 migration projection. That is, natural increase is negative in both cases. The total populations range from 5.7 million to 6.4 million. Thus, the six million theme used in this meeting is apposite.
This slide shows you the difference if the fertility rate remains around its present level. The main features are that the population is wider at the base, that is, it is younger and the range of the total population extends out to 6.8 million. Natural increase would remain positive at the end of the projection period.

This slide (below) summarises the total population from the four projections. All show considerable growth with a population of 5 million by about 2020 and 6 million by 2050.
Population of Sydney, 1999-2049
This slide (below) compares one of the Sydney projections, the lowest, with what is happening to the rest of Australia in a consistent projection (1.56 TFR and 90,000 ANM).

This is a projection for Australia that leads to the 25 million population in 50 years time and close to zero growth subsequently. The essential difference between the two is that the Sydney population is younger. The Sydney population is beehive-shaped and the rest of Australia is somewhat coffin-shaped. As we shift Melbourne, Brisbane etc from the right side to the left side, this impression would become very pronounced. That is, a projection that provides a reasonable outlook for Australia is the sum of high population growth in the existing cities with considerable ageing and labour supply decline in the non-metropolitan regions. We need more work on this and we shall be doing this as a component of the AHURI study of future housing needs.

This slide (below) summarises the point showing the total population trends.
Population Growth of Sydney and Australia, 1999-2049
The proportion of Sydney's population aged 65 years and over (below). The difference in migration leads to the differences between the lines of the same colour. The differences between the lines of different colours indicates the effects of different fertility. Fertility as expected and as already discussed has a larger impact that migration, but the central conclusion is that the variation in ageing across these projections is very small (22.8% to 24.8% aged 65+). A doubling of the aged proportion is inevitable.
Percentage aged 65 and over, 1999-2049
Here we have (below) the balance of births against deaths in the four projections. In the lowest projection, births exceed deaths around 2036. If fertility were to remain at 1.7 births per woman, births exceed deaths throughout the 50 years of the projection, but would cross in the next decade.
births and Deaths, 1999-2049
For many purposes, it is important to project the future number of households (below) rather than persons. It is likely that single person households consume more than half of the consumption of two person households.
That is, consumption is likely to rise as household size falls. There are also implications for housing policy of smaller households. The table shows that the average size of households is set to fall sharply in the coming two decades. Between 1996 and 2021, Sydney gains 980,000 people but 560,000 households. Thus the rate of growth of households is considerably faster than the rate of growth of population.

This (below) shows the change in the types of households over 25 years from 1996 to 2021. Unfortunately the data relate to NSW rather than to Sydney, but the picture for Sydney would be similar. The big growth occurs for couples without children and for lone person households - two and one-person households. Each of these groups increases by more than 300,000 households.
Together, these two household types would constitute well over half of all Sydney households. They are concentrated at the older and the younger ages, over 50 and under 30. However, there is still growth in the numbers of families with children, especially one-parent families.


The question is: where do we place half a million additional housing units in Sydney over the next 20 years and what types of housing should be provided. There are few greenfields sites left. The Badgery's Creek region has just become an option again and there is no better way to stop a future airport than to construct houses there. But there are environmental issues involved in filling up more of the already fragile Nepean basin with houses. Young one and two person households have already indicated a strong preference for inner city, medium and high-density housing. This group will probably spread out from the inner areas to the next circle as prices rise in the inner area.

Price rises will be affected by the growing numbers of temporary international migrants who have a strong demand for medium and high density inner city housing. So gentrification is likely to spread further out from the centre. We see this, for example, in the redevelopment of old factory sites along the Parramatta River. Getting rid of a few more dirty factories from the centre of the city is not such a bad idea - and almost certainly, inevitable.

A key issue is what the empty-nest couples will do. I have already speculated about whether they will leave Sydney altogether, but will they want to move into new, medium and high density housing in the inner parts of the city. If some do, that would free up houses for the families with children further out and perhaps obviate the need for major new greenfields development. However, I imagine most will stay in their houses in the suburbs that they know and love so that they remain close to their grandchildren, tend their gardens, participate in the local community and have space available for their children to return whenever their domestic arrangements fall apart. If this is the case, as the young generations in Penrith and Campbelltown, Gosford and Sutherland grow up, form relationships and have children, they will be looking for more three bedroom houses close to where they grew up - just as their parents did in moving from Parramatta to Penrith or from Hurstville to Sutherland. Where do they go?

Also, are we ready for the ageing of the population? Do we have sufficient housing suitable for older one- and two-person households in outer areas? Older people may wish to remain in their own locality but be unable to cope in their present style of housing. If they move out of their larger houses, these would become available for younger families. But some who are still a bit younger and a bit healthier may attempt to prevent older people being housed in their own locality because this would mean building new medium density housing in the locality. When it comes to providing suitable housing for older Australians, the NIMBY syndrome will be an obstacle.

Can Sydney grow outside Sydney? The paper I referred to earlier on Sydney's Future recommended that Sydney grow in Newcastle, not the whole extra two million but some of them. I like to keep an open mind but I would have to say that I am sceptical of this option as well. It doesn't fit the nature of the changing Sydney economy. To refer to Sassen again, the information age is promoting greater concentration rather than dispersion.

While I dislike coming to Sydney the city in which I was born and grew up because it is fast and congested, other people thrive on this kind of environment. Thus, while it all sounds horrendous to me, I suspect that Sydney planners have little choice but to plan for a future with six million people in Sydney in 50 years time. A master plan of infrastructure for this development would be a useful way to assess where problems are likely to arise. States now have their GST tax revenue base. This gives them greater potential to plan their own destinies.

In preparing this presentation, I have made use of the excellent reports prepared by the Victorian Department of Infrastructure. NSW could do well to emulate this work. But the comparison with Melbourne has left me with another thought and that is that the two cities should in future see themselves as being more complementary than competitive.

Melbourne is well placed to be the leading port for both sea and air transport. Its capacity to move goods is way superior to that of Sydney. It is well placed for whatever manufacturing that remains necessary in Australia. Sydney is not well placed for manufacturing or shipping of hard goods. It is primarily a financial and information centre. As Melbourne's fertility rate falls, perhaps some of Sydney's future population can be diverted to Melbourne rather than to Newcastle.

Peter Woods later today is talking about whole of State development. Maybe there is even more scope across borders. Ideas of a very fast train have fallen by the wayside because they are judged to be not viable from the perspective of the number of passengers carried - but what about very fast freight trains from Sydney to Melbourne? Or combined freight and passenger services? If the Cabinets of the two States can meet to talk about the Murray River - a great initiative - maybe they can talk about cooperative development on a much wider scale.
back to Population Forum 2001 index