Issues Magazine

Australia’s Future Population: Can We Forecast and How Should We Plan?

By Charlie Nelson

How are forecasts, targets and scenarios used to plan for population growth? Is bigger necessarily better when we look beyond GDP to measures such as life expectancy and happiness?

Australia’s population growth rate has increased dramatically since 2004. The current growth rate is the highest since at least the early 1970s, yet I cannot recall any forecast in the mid-2000s that predicted this growth rate. Nor do I remember any Federal government announcement that there was a target of 22 million by the end of 2009. Where was the plan for providing housing, transport infrastructure, education and health facilities for this population? In the area of housing, for example, the Australian Financial Review (28 April 2010) reported National Housing Supply Council estimates that there was a shortage of 178,000 dwellings in mid-2009 and that the deficit is still growing.

If we are not able to predict population growth five years ahead and we cannot provide basic needs for that population, what hope is there that we can plan optimally for the next 30 or 40 years?

In Australia we are fortunate that we can still choose to have a relatively small population. This option is not available to most countries with a large land area, and so we should very carefully evaluate the benefits and costs of a much larger population before we commit ourselves.

Forecasts, Scenarios or a Target?

As recently as late 2006 I did not forecast the large increase in our population growth rate. It was clear that both births and net migration were increasing, but the subsequent boom was not apparent. I predicted an increase of 1.3% in Australia’s population in 2007 (the outcome was 1.87%). I am not aware of any more accurate forecasts made at that time. The main cause of the greater increase than forecast was a rise in net migration from 182,000 in 2006 to 244,000 in 2007, and 301,000 in 2008 (the 2009 figure is expected to be about 300,000).

The 2010 Intergenerational Report (IG3), the third of its kind and released by the federal Treasurer in January 2010, includes a forecast for an Australian population of 35.9 million by 2050. This is much larger than predicted only three years earlier in the second intergenerational report issued in 2007 (IG2). The IG2 forecast for 2047 was 28.5 million, while the IG3 forecast is 34.9 million, an increase of 22%.

Signs are emerging that population growth is now slowing, perhaps challenging the view that Australia’s population will be as large as

36 million in 2050. Net migration in the June 2009 quarter was 10.8% lower than for the June 2008 quarter, the first such decrease since 2004. In the September 2009 quarter, net migration was 2.0% lower than in the September 2008 quarter. Births were down by 4.8% and 5.8% in the June and September quarters, respectively, compared with the corresponding quarters of 2008 – there had not been a decrease since 2006.

Population predictions are reasonably accurate in the one- to two-decade horizon, according to William A. Sherden in his 1998 book The Fortune Sellers. In this book he reviews seven forecasting professions, including economic and weather forecasters. His analysis shows that the track record of population forecasters is not as dismal as that of economic forecasters. However, of population forecasters, he concludes:

  • They have no forecasting “skill” because their predictions are only as accurate as those made by simple, naïve methods, such as using a ruler to extend past trends into the future.
  • They are unable to predict turning points in any population components (births, deaths, and net migration). For example, they missed both the baby boom of 1947 to 1961 and the baby bust of the 1970s.
  • Their predictions are subject to situational bias, as their predictions reflect only the trends existing at the time they made their forecasts.
  • They have made no progress in forecast accuracy during the decades since track records have been maintained.

If we cannot accurately forecast beyond 10 or 20 years into the future, and at times of rapid change in population components even shorter-term forecasts are fraught with error, how can we rely on a forecast of population size, age distribution and geographical distribution 40 years ahead?

Given the uncertainty about Australia’s future population size, another approach is to develop scenarios. These are plausible futures based on a range of possible trajectories for births, deaths and net migration. A set of three or four such scenarios would define an envelope within which we could be reasonably sure the actual population would lie.

Such scenarios are produced from time to time by the Australian Bureau of Statistics (which calls them population projections). Their most recent projections were released in 2008 and were based on 2006 Census data. The ranges of assumptions employed were:

  • total fertility between 1.6 and 2.0;
  • net overseas migration between 140,000 and 220,000; and
  • life expectancy at birth for males between 85.0 and 93.9, and for females between 88.0 and 96.1.

Further assumptions were made concerning internal migration in order to provide a range of projections by location.

The range of population size outcomes for 2050 was 30.2 million to 39.6 million.

Scenarios like these are valuable for planning, assuming that long-term plans are actually developed and implemented. The strategy is to develop plans that are sufficiently robust and adaptable to provide the services needed, whatever the outcome within the range. As time progresses, the scenarios should be updated and plans adapted.

Australia has never set a population target but is now discussing doing so. If we set a target for the 2050 population, how likely is it that we could achieve that target accurately? To do so would require the agreement of successive governments to control net migration (since they can’t control births and deaths) to within a tight band over a long period of time. This is unlikely given our system of elections every three years and an almost rigid focus on the short-term.

Identifying problems and opportunities decades ahead, then developing and implementing effective strategies, is not something that governments in Australia or other countries have done well. In the field of demographics, for example, it became clear by 1980 that the baby bust of the 1970s was a long-term problem that would lead to a rapid increase in the proportion of the population aged over 65 by the early 2010s. And yet,

30 years later, little has been done to address the problem. Similarly, it is nearly 20 years since the Rio Earth Summit at which nations agreed to take action to avoid dangerous climate change. Nothing of major consequence has been achieved since then, by Australia and most other countries, and there are no clear signs of significant action in the next three years at least.

Benefits and Costs of a Larger Population

Assuming that Australia is capable of a radical improvement in its long-term planning effectiveness, how should we now decide on the size of our future population?

As noted in the IG3 report, “population growth will also put additional pressure on infrastructure, services and the environment. Projected population growth is manageable, if governments plan for future needs.” It is easy to say that future problems can be managed if governments plan for future needs but they have little track record of doing so.

A much larger population would mean a much larger economy, as there would be higher productive capacity and more demand. But would we be financially better off per capita, and would we have a better quality of life?

The graph in Figure 1 plots population size and gross domestic product (GDP) per capita for the countries that the International Monetary Fund considers to be “advanced”. The key points illustrated by this analysis are that there are countries with larger populations than Australia but lower per capita GDP (Canada, UK, France, Germany, Japan) and there are also countries with smaller populations than Australia but higher per capita GDP (Norway, Singapore, Switzerland, Hong Kong). Clearly, population size is not correlated with per capita GDP. Australia is already one of the richest countries, but to improve per capita GDP we should look not to population growth but to improving productivity, workforce participation, and to innovation.

Are citizens of countries with large populations happier than citizens of countries with small populations? Apparently not. Australia is already one of the happiest countries.

Do countries with larger populations have longer life expectancy than countries with smaller populations? No, they don’t. Australia already has a higher life expectancy than many countries which are advanced economies and which have a much larger population, such as the USA, UK, Germany and Spain.

Both happiness and life expectancy appear to increase with GDP per capita, but only up to the level that which Australia has already reached.

Most Australian adults do not believe that Australia would be a better place with a much larger population, based on surveys by Foreseechange, each of which canvassed the opinions of 500 adults. Other polls also indicate that there is not majority support for a much larger population. For example, a Nielsen poll in November 2009 found that 40% of people surveyed thought that a population of 35 million by 2049 was too high, and 30% thought that it was about right (26% had no opinion and 2% thought it was too low). A 2010 Lowy poll found that 69% of respondents want the population to remain below 30 million.

Australians have not been convinced that a much larger population would result in a better country. Indeed, the analysis outlined above suggests that they are quite right to have doubts about the purported benefits.

A benefit claimed by some for higher immigration, and hence a higher population, is that this will tend to offset the rapid increase in the old age dependency ratio. This claim has little validity since a migrant aged 40 arriving today would be retired from the workforce by 2040, perpetuating the problem.

Inevitably, there will be many arguing that we should just get big, in terms of both population and GDP. But there is too much focus on GDP and it is not a particularly good measure of progress.

In February 2008, the President of the French Republic, Nicholas Sarkozy, created the Commission on the Measurement of Economic Performance and Social Progress to investigate how to improve the current state of information about the economy and society. The aims were to identify the limits of GDP as an indicator of economic performance and social progress; to consider what additional information might be required for the production of more relevant indicators of social progress; to assess the feasibility of alternative measurement tools; and to consider how to present the statistical information in an appropriate way.

The Commission’s report provides some compelling arguments for changing the way we measure progress:

  • GDP is an aggregate measure and provides no insight into the distribution of income and wealth. It is quite possible for GDP per capita to rise while most people are becoming worse off.
  • GDP and other commonly used official statistics may not measure factors that are important to the well-being of citizens. For example, worsening traffic congestion may be a positive for GDP in the short-term as more petrol is refined and sold. But the increase in air pollution and contribution to climate change may negatively impact quality of life. Furthermore, the congestion will negatively impact on productivity and so reduce future GDP growth.
  • GDP is an inadequate metric to gauge well-being over time, particularly with regard to environmental sustainability.
  • The global economic crisis of 2008 and 2009 has highlighted inadequacies of current economic measurement systems, which did not signal that the seemingly strong global economic performance between 2004 and 2007 may have posed some threat to future economic growth.

The single-minded goal of a larger GDP primarily via a larger population is unlikely to deliver any significant benefits to the community, but the costs are potentially very high.

The benefits of a much larger population have not been convincingly articulated and we would be better to become richer by being smarter and more productive. Given substantial doubts about the ability of governments to effectively plan for and deliver on time the infrastructure needed to avoid water shortages, environmental degradation, and to house and transport the population without worsening congestion, this would be doubly the smart option.

Choosing a Vision for the Future

Australia is fortunate in many ways. We have a very high standard of living and a resilient economy, we have a bountiful supply of many commodities, we are happier and healthier than many larger countries, and we still have the opportunity to decide what sort of country we want to be.

But we do not have a good track record for long-term strategy. The worst outcome would be that we stagger blindly into an accidental, irrevocable future.

Tony Blair, former Prime Minister of Great Britain, delivered an inspirational speech early in his term of office. He said that Britain could no longer aspire to be the biggest or the mightiest country. But, he said, Britain could be the best country.

Australia has a golden opportunity, for a short while, to determine what it means to be the best country, in terms of economic wealth, quality of life and sustainability. If we could agree on the right balance and commit to the necessary courses of action, we would indeed be the lucky country.