Immigration no cure for population ageing

ScreenHunter_2187 Apr. 28 08.37

By Leith van Onselen

In the wake of the Reserve Bank of New Zealand’s and New Zealand Treasury’s recent reports arguing against high immigration as it lowers living standards by reducing productivity, a new report has been released by Dr Katherine Betts from the Monash University Centre for Population and Urban Research, which claims that high immigration is an inappropriate policy response for population ageing, and can instead drain the nation’s productivity by crowding-out productive investment and capital deepening:

If we were to follow the stable path the population would grow from 22.7 million in 2012 to just over 27 million in 2066 and stay at around 26 to 27 million thereafter. The median age would rise from 37.3 in 2012 to around 47 in 2063 and then stabilise.

But so far we are not heading down this path. Instead governments have opted for immigration- fuelled population growth. From 2003 to 2012 the TFR has averaged 1.9131 but immigration has ballooned to record levels. The average (mean) annual net intake from December 2006 to December 2012 was 228,343…

In percentage terms the increase in population from net migration averaged 0.9 per cent from 1950 to 1969. From 2006 to 2012 it averaged 1.06 per cent. Thus the intake is not only now much higher in numerical terms, it is higher in percentage terms, and if it were to remain at this percentage level (or at any constant level in percentage terms) the nation would indeed be growing exponentially…

Figure 10 shows six of the projections from the 2013 series. All assume high life expectancy but the fertility and migration assumptions vary in such a way that the six projections form two sets. The three in the first set assume nil net migration (balanced migration — number of arrivals equals number of departures),135 but different levels of fertility. The three in the second set include one with a TFR of 2.0 and ‘low’ NOM while the other two both have high NOM and different levels of fertility. Comparing the two sets demonstrates that high levels of NOM, whether they be 200,000 per year or 280,000, make an enormous difference to the eventual size of the population. Indeed there is no end to expansion; the three projections in the second set are still growing briskly in 2101.

ScreenHunter_2188 Apr. 28 08.56

Australia is getting bigger but, if governments maintain these levels of immigration, are we going to get younger? Up to a point, yes. If we embraced the stable projection (series 56) the median age would rise from 37.3 in 2012 to 46.8 in 2061 and then stabilise between 47 and 48. But if we were to stay on the path mapped out by the high-growth projection (series 1A) the median age would rise to 43 in 2101 though it would still be increasing then, albeit slowly. High net migration does make us a few years younger: a median age of 43 instead of 47. But like most magic spells there is a catch. This is massive population growth, including many more older people. For example, in 2061 high growth would result in 48 million people. The size of the population aged 65 plus would also have risen from 3.2 million in 2012 to over 18 million in 2101. And the numbers of older people and the total population would both still be growing. (The stable projection has the numbers aged 65 plus in 2101 holding steady at 8.3 million.)

Figure 11 illustrates the consequences for the median age of each of the six projections shown in Figure 10. A comparison of the two graphs shows that, while the positive NOM series produce high growth, they make only a marginal difference to the median age. Indeed projection series 14 (TFR 1.6 and NOM 280,000) leads to a population of 45.4 million in 2101, but one that is marginally older than the stable projection of series 56 (and, of course, one that is still growing).

ScreenHunter_2189 Apr. 28 08.59

Projection series 68, which assumes a TFR of 1.6 and nil net migration is the one that produces the highest median age, 51.8 in 2061 and 55.5 in 2101. Assuming that we do not want to have a population with a median age as high as this, that we do want to avoid hyper-ageing, what is the most efficient way of arriving at a more youthful median age?

Table 1 takes series 68 as the benchmark and shows the relative effects of other fertility and migration assumptions on the median age in 2061. Table 2 sets out the same analysis for 2101.

ScreenHunter_2190 Apr. 28 09.01
ScreenHunter_2191 Apr. 28 09.01

In both Tables 1 and 2 it is clear that reducing the median age by one year via high migration is expensive in terms of numbers of extra people, with all their added pressure on infrastructure, cities, services and resources. By 2101 the high-growth series (1A), which assumes a TFR of 2.0 and a NOM of 280,000 per year, costs up to 4.1 million extra people per one year shaved off the benchmark age of series 68. In contrast, series 56 (the stable projection) costs only 0.98 million extra people per extra year of youthfulness. It is also a much more cost-effective method of reducing the median age than is projection series 14, with a TFR of 1.6 and NOM of 280,000. Series 14 leads to an older median age in 2101 than does the stable projection series but nonetheless adds 4.8 million extra people for every year shaved off the age of the benchmark series.

The message from the 2013 set of projections is clear. If policy makers genuinely want to minimise demographic ageing at the least cost, the most effective way of doing this is to support the two-child family and minimise net migration…

The Productivity Commission report on ageing points out that the infrastructure spending needed to manage population growth over the next 50 years will be five times the total that was needed over the last 50 years. This investment in capital widening must seriously weaken Australia’s capacity to invest in the capital deepening that would boost productivity.

Despite this, Treasury continues to emphasise its ‘three Ps’: population, participation and productivity. While Treasury treats these three variables as if they were independent some commentators argue that population growth has a positive effect on productivity. But there is a contrary argument. Population growth imposes pressures on infrastructure and adds to congestion; in so doing it depresses productivity.

International comparisons show that there is no association between population growth and growth in per capita GDP. This is not surprising as comparative data on 32 OECD countries show no positive association between population growth and growth in labour productivity…

Assertions that immigration-fuelled population growth will boost productivity remain conjectural. There is no empirical evidence that such growth in an advanced economy increases productivity. This means that advocates of population growth are left with the argument that it should be pursued in order to reduce the average age of the population. It may do so, to a limited degree, but adopting this strategy represents a considerable effort for a minor reduction in the median age. This costly benefit would also be fleeting. As no population can grow for ever, the median age of 47.5 would still be waiting for us when we slowed down…

In contrast the balanced migration route of the stable projection (series 56) would lead to a stable number of older people: around 8.3 million out of a 26.5 million. We would have turned 47.5 faster, but with much less stress.

As I keep arguing, a big negative of high rates of immigration is that it places increasing pressure on the pre-existing (already strained) stock of infrastructure and housing, reducing productivity and living standards unless costly new investments are made, which in turn chokes-off other productive investment.


Indeed, as explained in a 2011 speech by the Reserve Bank of Australia’s Phil Lowe (summarised here), rapid population growth (immigration) since the mid-2000s has placed upward pressure on rents, as well as caused a big surge in utilities prices as the capacity of the system struggled to keep pace with the growing demand, requiring costly new investments.

In a similar vein, modelling by the Productivity Commission has found that immigration is neither beneficial for the economy or living standards, nor can it sustainably alleviate the impacts of an ageing population.

Any objective examination of the facts suggests that the case for a high level of immigration is anything but clear-cut and those advocating a strong migration program need to justify their position.


[email protected]

About the author
Leith van Onselen is Chief Economist at the MB Fund and MB Super. He is also a co-founder of MacroBusiness. Leith has previously worked at the Australian Treasury, Victorian Treasury and Goldman Sachs.