By Colin Lynch*
I’m pleased that there is a high interest in the population projections Statistics New Zealand produces.
Informing public debate and decision-making is, after all, why Statistics NZ publishes information about New Zealand's population, economy, society, culture, and environment.
Andrew Whiteford's recent opinion piece makes some fair points about how different non-demographic factors can affect migration and population at both at a national and local level.
However, I do need to correct some factual errors in the article that are misleading in several respects.
The article refers to old projections, including Southland population projections published in 2007. Statistics NZ has published two updates since then, with both projections indicating population growth in the short-term and slowing growth in the long-term.
For NZ population projections, Statistics NZ no longer publishes "nine different scenarios", but a more comprehensive set of probabilistic projections with indications of uncertainty.
The latest projections can be accessed from: www.stats.govt.nz/estimates-projections
When we talk about future populations, it is natural that things will change, which is why the projections are up-dated so regularly.
The projections do not try to anticipate major policy changes. Changes to the Immigration Act in 1986-87, for example has had a major bearing on migration trends.
It’s for this reason that the projections are not forecasts or predictions.
Sometimes projections will be higher than what occurs; sometimes they’ll be lower. Because the future is uncertain, Statistics NZ never publishes a single projection.
The projections are an indication of future population size - and more importantly, of population age composition - and users should always use the latest projections.
The article encourages incorporation of "economic prospects" into population projections, although it's not explained how this might explicitly be done for projections out 10, 25, or 50 years. That this is not done by other international statistical agencies indicates that the requisite economic projections simply aren't available.
Fundamentally, only births, deaths, and migration can change our population, and given the complex relationship between economic factors and migration, it perhaps make sense that population projections are based on those three components.
The criticism of life expectancy assumptions ignores Statistics NZ's considerable and published work in this area.
Firstly, the article overstates past life expectancy gains over a period of more than 20 years, and understates the latest assumptions.
Secondly, life expectancy is a summary indicator of the death rates at all ages, and it’s trends in these death rates which drive the assumptions used in the population projections. Readers interested in a full discussion of life expectancy trends are encouraged to read our projection release and 'Forecasting mortality in New Zealand' paper: www.stats.govt.nz
The value of statistics lies in them being analysed and used, so it is good to see such strong interest in the quality information produced by Statistics New Zealand.
---------------------------------------------
Colin Lynch is the Deputy Government Statistician at Statistics New Zealand
We welcome your comments below. If you are not already registered, please register to comment.
Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current comment policy is here.