International migration and the growth of households in Sydney: a comment.

The previous issue of People and Place contains a paper by Peter McDonald and Jeromey Temple which analyses house-hold formation in NSW under various assumptions regarding Annual Net Migration (ANM). (1) The declared purpose of the paper is to 'draw conclusions from the results of some recent housing demand projections that we have completed'. (2) To this end a newly-developed model called Ozhouse is applied to regional housing demand projections from the year 2001 to 2031. Five alternative Scenarios or Series are postulated with varying rates of international and internal migration in and out of NSW in general and of Sydney in particular.

The lowest rate of international migration considered in the paper is zero net. This is outlined as Series 4 and unsurprisingly yields the lowest rate of population increase and household formation of the five alternative scenarios, particularly in Sydney (Figures 3 and 5). The results in these Figures indicate that international immigration is a major cause of housing demand in Sydney, but this is not the conclusion drawn in this paper.

The paper describes the Series 4 zero net migration series as 'a hypothetical counter-factual'. (3) The results of the Series are largely sidelined or disregarded by the authors. The purpose of the present comment is to question four assertions made in the paper and suggest that an alternative concluding remark would fit the authors' data extremely well.

Table 1 in the paper outlines annual net migration, international and internal combined, for Sydney and the balance of NSW, in 2001 and 2030, according to the five migration scenarios. It shows that under Series 4 Sydney would have an annual loss of about 12,000 people while the balance of NSW would grow by about 20,000 per annum. Thus zero net migration would reduce pressure on Sydney while providing a population boost to the regions, a result that many would regard as highly desirable. However McDonald and Temple exclude Series 4 from consideration on the grounds that it is 'very hypothetical'. (4) This is a curious argument since all projections are based on hypothetical assumptions, and no explanation is offered as to why this assumption is more hypothetical than any other.

McDonald and Temple claim that 'on present trends, the imbalance between projected numbers of aged people and workers in coastal areas of NSW is unsustainable'. (5) This statement is not substantiated and appears to be based on a general fear of an ageing society, a fear which has often been shown to be exaggerated. (6)

The population pyramid for Series 4 (Figure 3) shows a drop in numbers at younger age groups by the year 2031, so that 'the numbers entering the young adult ages would shrink even more rapidly after 2030'. (7) The authors describe this situation as 'a very unlikely and very undesirable outcome', (8) without substantiating either viewpoint. Given the present political climate the result may be unlikely in Australia but it already pertains in other countries such as Japan and Spain, both of which nations have had low fertility rates for some years, so the result is presumably not impossible in Australia. The authors' claim that the result is undesirable may reflect their own views, but a demographic outcome which could lead to a reduction in population size has many arguments in its favour. Indeed, it is often claimed that a fall in numbers is imperative if future generations are to enjoy a comfortable standard of living.

Figures 6 and 7 graph the growth in Sydney of households with persons aged 60+ and indicate that this rate of growth varies little with the different immigration scenarios. The authors comment that 'the ageing of the population is driving much of Sydney's future household growth'. (9) ...