ABUL RIZVI. Possible population projections in the forthcoming 2020 Intergenerational ReportMar 17, 2020
Treasurer Frydenberg will shortly publish the 2020 Intergenerational Report (IGR) –Australia’s fifth such document. It’s as close as we get to a long-term plan (usually 40 years) for the Australian population and economy.
So what can we expect will be in the plan? And how can we judge if it is anything more than just a political document such as the ten year plan Frydenberg published just before the 2019 Election or the so-called Population Plan published by Scott Morrison in March 2019?
Scott Morrison’s Population Plan infamously provided zero information on the Government’s views on Australia’s future rate of population growth, natural increase, fertility, mortality or net overseas migration (NOM).
A month after the Morrison Population Plan, Frydenberg published his ten year budget plan which did include, hidden away in Appendix A of Budget Paper No 3, the Government’s views on Australia’s future fertility rate, rate of natural increase and rate of NOM. These assumptions are crucial to any long-term forecast of Australia’s economy and budget.
But Frydenberg’s ten year plan made no mention of the impact of population ageing on the economy or the budget. Will his 2020 IGR consider the impact of population ageing?
Fertility and Natural Increase
Frydenberg in 2019 considered Australia’s fertility rate would rise to 1.9 babies per woman by 2021. The predicted increase in fertility meant Frydenberg could also predict that Australia’s rate of natural increase (ie births minus deaths) could stop declining and in fact begin to rise.
An extraordinary prediction given Australia’s fertility rate has been in steady decline for more than a decade (see Chart 1). In 2018 it was 1.74 babies per woman, the lowest level in our history.
Since 2008, Australia’s fertility rate has been declining rapidly, as has been the case in many developed economies as different as South Korea and the USA.
So what fertility assumption will Frydenberg use in his 2020 IGR?
With creation of a ‘population unit’ in the Treasury Department, Frydenberg will hopefully get better advice and retreat from his 1.9 babies per woman assumption.
An assumption of 1.8 babies per woman (reflected Series B in Chart 2) would be the very highest he may be able to justify if he can point to new policies that better support families with new born children (eg improved approach to reducing costs of childcare that won’t get swallowed up by childcare companies).
In all probability, Australia’s fertility rate rising to 1.8 babies per woman is highly unlikely. A more plausible assumption would be a fertility rate at around current levels (ie 1.74 babies per woman) or perhaps a steady decline to 1.65 which is the low fertility assumption used by the ABS in its latest population projections. This would still be above the OECD average fertility rate of 1.6 babies per woman.
A fertility assumption of 1.65 babies per woman would result in a steeper decline in the rate of natural increase (see Series C in Chart 2), particularly when combined with a sharp increase in the annual number of deaths in the period 2030 to 2050 associated with the passing of the baby boomers. Depending on the assumption used for NOM, the rate of natural increase would approach zero in the second half of this century and less than zero before the end of this century.
Negative natural increase is now standard for countries in Europe and Japan.
Net Overseas Migration (NOM)
Due to the economic impact of the coronavirus, NOM in 2020 will inevitably fall to very low levels. Given that in previous recessions Australia’s rate of NOM has fallen to around or even below 50,000, a very sharp fall in NOM should be expected. Immediately after the GFC, NOM fell by around 100,000 but off a relatively higher base. A fall of similar magnitude or even more would not surprise.
The key question for the 2020 IGR is what level of NOM should be assumed under current policy settings once the impact of the coronavirus has passed?
Despite the reduction in the permanent migration program, the 2019 Budget forecasts that NOM will rise strongly. This is after forecasting NOM would fall in the 2018 Budget (see Table 1.1).
For the period 2018 to 2022, Frydenberg increased the forecast of NOM by a total of 197,800 compared to Morrison in the 2018 Budget. But given the size of the cuts to permanent migration, is this plausible? Professor Peter McDonald has shown that that over time there is a strong link between the level of permanent migration and NOM. He shows that differentials between the two do not last long.
So is the Government aiming to break that relationship? This is where we need to be able to test the Government’s forecast for NOM but without the impact of the coronavirus.
In the six months to December 2019, we saw a sharp fall in offshore student visa grants, particularly from India and Nepal – Australia’s second and third largest student source countries. In recent years, students have represented over 40 percent of NOM.
There was also steady decline in the stock of working holiday makers and skilled temporary residents continuing in the six months to December 2019.
The contribution of permanent migration to NOM is also likely to decline, mainly due to the large reduction in the skilled independent category – a category that disproportionately grants visas to people who are outside Australia.
Australia’s high unemployment rate compared to that of New Zealand has ensured that New Zealand citizens make a very small contribution to NOM compared to the period before 2013.
Indeed, the only aspects of NOM that appear to be increasing are people arriving on visitor visas and then applying for either asylum or a partner visa. There has also been a decline in returns and removals thus increasing NOM but this is unlikely to continue given the rising number of unsuccessful asylum seekers in Australia.
Drawing on the above, it is likely current policy would result in long-term NOM in the range of 175,000 to 225,000 per annum – significantly below the average of 268,000 forecast in the 2019 Budget.
Series B in Chart 3 reflects a fertility rate of 1.8 and NOM of 225,000 per annum. It would result in the population growth rate steadily falling from 1.7 percent per annum (or over 400,000) to 0.8 percent per annum (or lessd than 350,000) by 2066. Series B would be the very highest rate of population growth Frydenberg could possibly assume for the 2020 IGR under current policy settings.
Series C in Chart 3 assumes a fertility rate of 1.65 and NOM of 175,000 per annum – that is far more plausible under current policy settings. It would result in the population growth rate falling to 0.5 percent per annum (less than 200,000).
In terms of population ageing, Series B would see the portion of the population 65+ rising from 15.6 percent in 2018 to 20.9 percent in 2066. Series C, with its lower fertility rate and lower NOM, would result in the portion of the population 65+ rising to 23.0 percent.
But what would slower population growth and more rapid population ageing mean for economic growth?
This will be addressed in a follow-up article on the 2020 IGR.
Abul Rizvi was a senior official in the Department of Immigration from the early 1990s to 2007 when he left as Deputy Secretary. He was awarded the Public Service Medal and the Centenary Medal for services to development and implementation of immigration policy, including in particular the reshaping of Australia’s intake to focus on skilled migration. He is currently doing a PhD on Australia’s immigration policies.