Sunday, January 28, 2007

Demographic Change in NSW - the future

In my Australia Day personal reflections I suggested the NSW Government's coastal strategies were flawed, perhaps dangerously so. I also provided a heads-up on the thinking that I had been doing.

This post extends my analysis. I do not pretend that this is definitive or even totally rigorous. I am simply trying to present a different view from that built into the official thinking. This comes both from my own personal biases - these cause me to ask different questions - and my insatiable curiosity.

So when I came to look at the NSW coastal strategies, one thing that stood out to me, one that I have been mulling over since is just why nobody, or at least nobody that I am aware of, has been looking at the overall patterns of development and change in the current state, starting with the demographic base, that must provide the real underpinnings for any strategies.

Macro View: Are the Projections Achievable?

The Government's various strategies suggest, if my maths are correct, that the population along the NSW coastal strip will increase by an average of 68,000 people per annum over the next twenty-five years, adding 1.69 million to the State's population, bringing that population to 8.49 million.

Are these projections achievable?

In 2005-2006, NSW gained 40,492 from natural increase (births minus deaths) , lost 23,970 through internal migration to other states and territories, but gained 42,231 from overseas migration for a net gain of 58,753.

There is a significant gap between this 58,753 and the 68,000 projected in the plan for the coastal strip. How might this gap be filled?

The pattern shown by the 05-06 numbers of gains from overseas migration partially offset by internal migration has been in existence for some years. It means that a 68,000 annual average increase in State population requires some combination of an increase in the rate of natural increase and/or some reduction in the rate of internal migration and/or some increase in the number of overseas migrants settling in NSW.

It seems to me that these outcomes are very uncertain.

Start with the natural increase. There has recently been a small but significant increase in the Australian birth rate. If maintained, will this be sufficient to increase the absolute size of natural population increase?

I think not. I am not sure how many people realise that the number of Australian women in the under thirty age cohorts is significantly less than the numbers in the over thirty cohorts. This smaller number of women will have to have significantly more children if the absolute size of natural increase is to rise. Further, and I have not calculated the effect of this, the absolute number of deaths will rise steadily over the twenty year planning horizon as a consequence of aging, adding further pressures.

What about internal migration? Will this fall?. Here I cannot see the rate of internal migration to other states and territories dropping simply because it the pattern seems so well entrenched.

All this means that migration to Australia remains (as it has been for many years) the key variable affecting NSW, predominantly Sydney, population growth. Here we strike a further set of problems.

Immigrant Numbers

To amplify this, take the 2005-2006 migration stats.

In that year, there were 131,593 settler arrivals, people migrating to Australia with permission to stay permanently, an increase of 7.5 per cent over the previous year. This was the headline figure reported in the media. But in that same year 67,853 permanent residents emigrated from Australia, an increase of 8.4 per cent. So Australia gained 131,593, lost 67,853, for a net gain of 63,740.

In addition to the settler arrivals, there are a substantial number of people coming to Australia as non-permanent residents - refugees and students - who may later become permanent residents. Once this happens they are added to the migration and main population numbers.

In 2005-2006, NSW made a net population gain of 42,231 from overseas migration. Of this, 19,325 came from those already here achieving permanent residence, leaving a net gain from settler arrivals/departures of just 22,906. Not a large number given the size of the state's population.

I must say that I had not realised the significance in statistical terms of the numbers of people already here achieving citizenship. So we now have three sets of variables - the number of people aready in NSW on a temporary basis who might achieve permanent residency and thus appear in the population numbers; the number of future people entering NSW on a temporary basis who might then achieve permanent residence; and the net number of new settlers (arrivals minus departures).

I think that the bottom line in all this is that we cannot actually assume that net international migration will yield the numbers built into the coastal strategies.

Internal Migration within NSW

There is another way the coastal strip could gain population, continued internal migration from inland to coastal areas. This has in fact been a pattern since the 1900s.

Leaving aside the question of desirability, there is actually an issue as to the extent to which this is either possible or likely over the 25 years covered by the coastal strategies.

Achievement of an annual average increase in the coastal population of 68,000 over twenty five years requires the coast to absorb all of last year's total state population growth from all sources including all migrants and the natural increase in inland NSW plus a further 9,000 per annum.
If this 9,000 were to come from internal migration from inland NSW, it implies a decline in the inland population over the period of 225,000 people, more than a quarter of the existing population.

I don't think that this is possible. If anything, my feeling is that the NSW population dynamics are shifting such that the inland population is actually likely to increase. To understand this we need to look at the evolving population pattern in inland NSW.

Inland Development

From my viewpoint, one of the interesting things about NSW is the way the state is evolving into two very different populations strips, one along the coast, the second along the Tablelands, Slopes and immediate Western Plains.

To illustrate this, lets start in the south east of the state. Here we begin with a tiger called Canberra and the ACT, a tiger whose influence is concealed by the myopia created by state borders.

Canberra is by far the most successful decentralisation experiment in Australian history.

At the end of June 2006, the ACT had a population of 329,000. Its growth has affected surrounding NSW areas for more than a 150 kilometres. The NSW city of Queanbeayan just across the ACT border has grown from perhaps 12,000 when I moved to live there in 1972 to 38,000 today. This growth continues to spread, with increasing number of ACT people moving to further out to Yass whose population has grown from a few thousand to thirteen thousand today.

The story does not end here. In South Easten NSW we also have Albury (41,000 plus those in Woodonga just across the boarder), Wagga Wagga (around 58,000), Goulburn (27,000), a little further out Griffith (over 15,000). The pattern of rise and fall in populations across this area is varied, but the total population including the ACT is now large enough to affect overall NSW patterns.

Moving further north to the central west we have Dubbo (40,000), Orange (34,000), Bathurst (33,000). While the bigger centres are a little smaller, they have been growing as have smaller centres such as Mudgee.

Further north we move into New England, my home territory and an area of deprivation whose relative influence has been declining for one hundred years. Even here we can see the same pattern in Tamworth (around 35,000), Armidale (22-25,000 depending on definitions), Inverell and Moree (each around 11,000). These centres have grown to the point that they are attracting people from elsewhere.

This total inland strip has a life increasingly independent of Sydney influences. Armidale, for example, has more people living within an eight hour car drive than Sydney. Road traffic between the major conurbations of Melbourne and Brisbane goes through inland NSW. Canberra growth spreads across a broad region.

All this makes me feel that inland NSW is at the cusp of a major change point and that, consequently, net migration from the inland to the coastal strip is unlikely to add to coastal populations in the way it has in recent years.

Migration and Changing Population Structures

We can look at the population statistic in another way.

For the sake of simplicity, assume that the 05-06 numbers continue for the next ten years.

During this period, NSW will:

  • add 405,000 people from natural increase
  • add 422,000 from overseas migration
  • lose 240,000 from internal migration to other states or territories
  • for a net gain of 587,000, bringing the total state population to 7,387,000.

All very simple, isn't it? But look again:

  • if internal migration doubles, population growth almost halves
  • if net oveseas migration to NSW halves, the NSW population growth turns to zero

Assuming that all the figures hold, the overseas born share of the NSW population will rise by a further 2.5 per cent. Because Sydney is the port of destination for most overseas migrants, the percentage rise in the overseas born in Sydney will be greater.

Now where will these migrants come from, what might this mean for planning purposes?

If we look at the break-up of the total 05-06 Australian numbers for settler arrivals, we find that:

  • there were 59,507 in the skills stream
  • there were 34,771 in the family stream, people coming to join relatives.
  • there were 12,113 in the humanitarian stream
  • with a further 25,098 non program arrivals, mainly Kiwis.

If we look at some other features of national settler arrivals we find:

  • 48.1 per cent (around 63,296) of settlers reported an occupation prior to settlement, 41.8 per cent were not in the work force (presumably children, partners) , 3.1 per cent were unemployed.
  • of the settlers reporting an occupation, 42.4 per cent (26,838) cent were professionals, 12.6 per cent (7,975) trades, 11.6 per cent (7,342) associate professionals. The rest - 21,141 - are not identified but are presumably unskilled.
  • of those in the humanitarian program, 52.1 per cent came from North Africa and the Middle east, 27.2 per cent from Sub-Saharan Africa, 12.9 per cent from Central Asia, 6.3 per cent from SE Asia.
  • Overall, the three largest source countries for NSW were the PRC 11.9 per cent, India 10.3 per cent and the UK 10.2 per cent.

Now what might all this mean for NSW planning?

First and perhaps least important, note the relative importance of New Zealanders in the overall numbers. This makes New Zealand developments quite important in terms of overall migration numbers, increasing the uncertainty attached to the numbers.

The second thing, one that I found interesting and even surprising, was the relatively large number of new settlers not in the workforce. This means that migration is contributing less than I had expected to work force growth, while adding to demands in areas such as schooling.

Now look at the different streams because each has different settlement implications.

Professionals and para-professionals appear to dominate the skilled stream. This is also the area where Australia has the greatest immigration, so the net effect is likely to be much smaller than the absolute numbers. If these groups behave like their Australian counterparts, then in locational terms, the NSW share is likely to be concentrated in particular parts of Sydney such as the Eastern Suburbs with the rest fairly evenly if thinly spread, while the trades component will tend to go where the work is.

The family reunion stream represents chain migration and will go in the first instance to where the families are. In the NSW case this mainly means certain parts of Sydney. This also holds to some degree for the non-program, mainly Kiwi arrivals.

The final stream, the often unskilled humanitarian stream, spreads much more widely because of the impact of Government policies. We can see this clearly in the debate about Tamworth Council's decision on refugees since this revealed that the Government's refugee settlement program is consciously settling humanitarian refugees such as the Sudanese in a wide variety of locations.

Migration and the NSW Government's Coastal Strategies

Now look at all this in the context of the NSW Government's coastal strategies, assuming that the basic population growth is there. Before doing so, I need to add one more piece of demographic data.

Sydney's population especially in the suburbs ringing the city from the south west to the north west is younger than than the national average and hence has a higher number of births. Conversely, with some exceptions the population elsewhere along much of the coastal strip is older than the national average and hence has a lower number of births.

Given this, the strategies seem to imply that the coastal strip outside Sydney, and especially the mid North Coast, will attract substantial numbers of new migrants from elsewhere. Since the majority of these are unlikely to come from overseas migration for the reasons given above, this means attracting locally born from elsewhere in the state and this means especially Sydney.

Will this happen and, if so, to what degree? I am very cautious.

I can see further retirement movements, although the net effect here will depend upon the balance of births and deaths given aging populations in retirement areas. By the 2032 deaths of current baby boomers will be having a very major impact in these areas.

I can also see some continued movement of people squeezed out of Sydney by that city's high prices. However, if these and any others who move are not to simply add to the pool of unemployed, there have to be jobs. Here all the strategies are very weak on job creation, assuming that the jobs will be there. I think that the Mid North Coast and Far North Coast strategies are especially vulnerable.

If we now look at Sydney, the Sydney strategy postulates that between 420,000 and 480,000 of the extra 1.2 million people projected for the conurbation over the next 25 years will be located in new urban developments in south-western and north-western Sydney. The balance of the population will be accommodated through more intense occupation of existing areas. The plan also provides for the further development of ring cities based on centres such as Liverpool or Penrith, each replicating in some way CBD style developments.

Will all this happen and, if so, what might it mean? Here I really do not know enough to make firm judgements.

I can certainly see more intense occupation of existing areas as a consequence of skilled migration and family reunions. As indicated, professional migrants are likely to cluster in areas already attracting professionals, while family reunion migrants and, to a lesser degree, humanitarian refugees will go in the first place to existing migrant suburbs where their kin or ethnic groups already reside. The second is likely to reinforce existing ethnic and migrant patterns.

I can also see some flow over into new suburbs as a consequence of pressures in existing suburbs. My primary concern here is the need to avoid the pattern of recent decades in which at least some outer new suburbs become centres of isolation and social deprivation. To the degree those suburbs are also marked by enduring ethnic differences, then we risk replicating Paris.

In the midst of all this, I am also fascinated by what it all means for Sydney's internal dynamics. We have already seen the city break up into a series of very different sub-cities that rarely interact. We have also seen waves of development progressively transforming and re-transforming individual areas. Social, economic and ethnic differences between different areas within Sydney are now very substantial. So we are going to see some interesting changes over the next 25 years.?

Migration and Ethnic Divergence

All of this has some interesting implications for population patterns across NSW.

Starting with NSW as a whole.

It seems to me that the current drivers of NSW population growth - natural increase, internal migration to other states and territories, the size of overseas migration - will continue their current pattern. This means that the size of overseas migration will continue to be the main driver. The aggregate population projections assumed in the coastal strategies will be achieved if and only if overseas migration levels increase to the extent required to bring the necessary population growth about.

If population growth continues just at its current levels and composition, the proportion of overseas born in the NSW population will continue to rise as a consequence of the combination of of overseas and internal migration, while the proportion of non Anglo-Celts in the population will continue to rise. The greater migration, the faster population growth, the greater these trends.

These aggregate changes conceal major distributional differences within NSW.

Sydney's younger population compared with other parts of NSW means that the city's natural increase will be higher in absolute and relative terms to the rest of the state. However, migration from Sydney to other parts of NSW as well as other states and territories will, as it has for at least the last twenty years, continue to be concentrated in the Australian born group. This will probably at least halve natural population increase.

The professional, and family reunion migration streams will continue to favour the city. Humanitarian migration will be more widely spread. This means that the non Australian born, non Anglo-Celtic proportion of Sydney's population will continue to rise faster than the state or national average. The greater our migration, the greater the impact.

Without going into details because this opens a complete new topic, there are considerable variations in the ethnic composition of the migrant intake between different cities and regions within Australia. The Asian feel of Sydney as compared to Melbourne's European feel in part reflects continuing differences in the composition of migrants. I think it likely that by 2032, the end of the 25 year planning period, the two cities will feel as though they are in different countries. Some would say that they already do!

I should add that I see nothing necessarily wrong with this. I love Melbourne's European cafe feel, but also like Sydney's Asian feel. The growing diversity should simply add to the Australian experience.

Internal migration from inland NSW including the ACT to the coastal strip will decline and may even go into reverse. Inland NSW will receive a proportion of the humanitarian and, to a much lesser extent, professional and skilled, migration, leading to a more diverse ethnic structure. However, the diversity gap between Sydney and inland NSW will continue to widen.

The coastal strip outside Sydney is the big unknown. Just because it is coastal does not mean that it is the same. In fact, I would expect very varying patterns.

The south coast south of Milton has always been the ACT-Monaro playground. Given the growth in the ACT and surrounding population, I would expect a flow of inland retirees to boost population over the next twenty years. I would also expect the Anglos-Celtic proportion of the population to remain higher, the migrant proportion lower, than the national or state average.

The lower Hunter and, to a lesser degree, the Illawarra have always had economic activity independent of Sydney. Newcastle in particular has its own hinterland and its own port. While the city and the Lower Hunter are now increasingly treated by Sydney as part of Sydney, they are very different. I would expect the area to grow of its own accord. In doing so, I would also expect the area to attract a higher proportion of migrants.

The Mid North Coast and, to a lesser extent, the Far North Coast are the biggest question marks. Yes, there are non-tourism growth hot spots such as Taree, but the overall job creation process appears far lower than that required to support the population growth numbers.

No comments: