I’m pretty annoyed about the comments Maurice Williamson made last week. In his capacity as Minister of Statistics, he has taken data produced by Statistics New Zealand, and misrepresented it. I know I’m being idealistic, but I don’t think statistics should be treated that way, and certainly not by the Minister. Quoting the Herald from earlier this week – and he also made similar comments on various news channels:
“Statistics Minister Maurice Williamson… declared the first Census data in seven years contained a surprise “bigger than Ben Hur”.
The finding that so enraptured the minister was that New Zealand’s population growth had halved since the last Census. The population had increased by 31,000 a year, or 0.75 per cent, over the past seven years, compared to 58,000 a year in the previous 2001-2006 period.
But this wasn’t news. As mentioned on the blog, the slower population growth is in large part due to lower migration (plus some other factors which I’ll look at below). It didn’t come as much of a surprise at all, to anyone who’s familiar with the information coming out of Statistics New Zealand. It’s possible that the Minister of Statistics is not included in this group!
“Net” International Migration into New Zealand (Immigration minus Emigration)
12 Months to February | Net Migration | 12 Months to February | Net Migration | |
2002 | 21,992 | 2007 | 13,151 | |
2003 | 41,557 | 2008 | 4,643 | |
2004 | 30,075 | 2009 | 6,160 | |
2005 | 11,131 | 2010 | 21,618 | |
2006 | 8,274 | 2011 | 8,249 | |
2012 | -4,068 | |||
2013 | 1,195 | |||
AVERAGE | 22,606 |
| AVERAGE | 7,278 |
At the national level, at least, Statistics New Zealand have got a pretty good idea of how the country’s population changes over time, thanks to records of births, deaths and international migration.
These figures are released on a regular basis, and Statistics New Zealand uses them to produce national population estimates on a quarterly basis. They’re released without much fanfare, and they don’t tend to make headlines, but plenty of people rely on them. The estimates get rejigged every five years, when census data comes out, as that is considered to be the most accurate source of population information.
As at 31 March 2006, New Zealand’s estimated resident population was 4,176,100. This is 3.67% higher than the “usually resident population” figure from the 2006 census of 4,027,947 people. The difference reflects census undercount and another few weeks of population growth (the census was held on 7 March 2006).
As at 31 March 2013, New Zealand’s estimated resident population was 4,463,900. This estimate will be revised in the future, but let’s look at the figure as it currently stands, and see if the Minister’s claims stand up to scrutiny.
The estimate is 5.23% higher than the “usually resident population” figure from the 2013 census of 4,242,048 people. The same factors are in play here as in 2006 (census undercount and a few weeks of population growth). So it’s possible that Statistics New Zealand had slightly overestimated the population, and those estimates could now be revised downwards slightly. Emphasis on the “slightly”.
Regardless, the difference between what we’ve just counted in the census and those population estimates is hardly “bigger than Ben Hur”. And there are some other issues as well.
New Zealanders travel a bit more these days, so there would have been more people overseas on census night (this is shown up in “average number of short-term NZ travellers overseas by purpose” statistics) – in March 2006, there were an average of 62,928 Kiwis overseas at any one time. In March 2013, the figure was 78,125.
However, I think it’s also quite likely that the non-response rate of the census has increased. My understanding is that statistical authorities are dealing with this issue around the world – even a quick Google brings up articles like this one, suggesting that countries were achieving response rates of around 98% in the early ’90s. Today, it’s more like 96%: Statistics New Zealand estimate that they achieved 96.2% coverage in 2001, and 96.0% in 2006. On current release schedules, Statistics New Zealand won’t let us know their latest estimate until early next year. But the general trend internationally has been towards lower response rates.
Overall, New Zealand’s population seems to be more or less in line with the estimates that were already in the public domain, and Auckland’s population growth has also been similar to what those estimates say (at more than 50% of national population growth).
Conclusion
So, in conclusion, it’s very misleading for Maurice Williamson to say that population growth has declined significantly and then act as if this is a major surprise. The people of New Zealand deserve better from the Minister of Statistics. So do the staff at Statistics New Zealand, many of whom I’ve dealt with over the years. These kinds of statements can undermine confidence in the statistics. They could even undermine investment, because investors think that it’s just been discovered that population growth is way lower than everyone thought. As I’ve written above, though, that’s not the real story.
Anyway, that’s enough of that. There will be a bit more census info released today – population counts all the way down to meshblock level (these are often similar in size to a city block, so pretty detailed). The more interesting stuff will start to come out over the next few months, and it’s good to see that Statistics New Zealand seem to be running ahead of schedule with the census releases. We’ll bring you more coverage as new information is released.
Processing...
It’s pretty clear that that announcement was politically motivated being an attempt to affect the local body election outcome. Further evidenced by the the fact that he only picked up on two issues that National are opposed to i.e. the compact city and a rail link is that would be affected by this supposed ‘surprise’.
This government has turned misquoting stats into an art form..You see we are all really stupid (the polls over the last few years prove it, they reason) and we won’t notice their cleverness..
Do you mean like how ACC made a huge loss in 2009 but in 2013, there is a huge profit? All due to ACC’s investments, and the effects of the GFC, but the Bill English wont tell you that. The average person just thinks ACC was run poorly and now it isn’t.
And levy rises galore don’t forget.
And cutting ACC payments, including for those people with pre-existing conditions like pregancy. That saved a dollar or two too, I bet.
If Williamson genuinely understood what these statistics indicated then I would expect him to comment with just as much surprise regarding the reduced road traffic growth over the last several years.
Using his same logic he would undoubtedly conclude that a number of future road building plans were probably unneccessary.
Should I hold my breath?
good post John – you have perfectly captured why Williamson’s comments on population growth in general, and the implications for the CRL in particular, were so uninformed.
It’s a real shame (and a concern) when Ministers such as Williamson simply rely on a tired ideological agenda (i.e. pro-motorways; anti-rail) rather than considered and intelligent analysis.
“Its a real shame (and a concern) when Ministers such as Williamson simply rely on a tired ideological agenda (i.e. pro-motorways; anti-rail) rather than considered and intelligent analysis.”
Thats because ministers like Williamson, Joyce, Brownlee et al are neither considered or intelligent.
Data coming out from Stats NZ today shows that Auckland’s population has grown by 110,000 to 1,420,000 over the last seven years. That’s 8.4% growth over the seven years or approx 1.2% growth per annum. Doesn’t sound too bad to me.
The Herald had a good interactive graphic on their website today that lets you compare population changes at the Territorial Authority Level and the Mesh block (I think) level.
You can access it at this url: http://www.nzherald.co.nz/nz/news/article.cfm?c_id=1&objectid=11140576
[May need Java installed to use the interactivity on this page].
Using this you can see percentage population change across NZ and within Auckland, Wellington, Christchurch Local Authority areas.
[there is a slider bar to let you see the “previous” census graphed and then this census to compare like for like].
You can also download the actual Census Data they used as a workbook for the above from the web site (Public Tableau)
– download and install the free tool from this site, then download and open the “NZ 2013 Census workbook” from the same link in the tool to slice and dice the figures).
http://public.tableausoftware.com/download/workbooks/NewZealandCensus2013Population?format=html
This site may provide a Windows only version [it came to me in a msi file which means its my download was Windows only], not sure if its got other versions there too.
I had a quick play and the tool is quite nice, allowing you to see the actual underlying data for the last 3 census (2013, 2006 and 2001) and compare.
Throws up some interesting stats like these ones:
Stonefields (as expected), growth was 936% from 2006 to 2013, and now has 2,238 living in it. as compared to 156 back in 2001, and 206 or so in 2006.]
However, Stonefields is not the fastest growing area in the 2013 census, (its only the second fastest)
– Mission Heights takes that prize at 5175% growth, for a population of 2,532 in it.in 2013.
Average population density in Auckland TLA per Km2 is 2,213 (averaged over the 410 meshblocks)
Which is not particularly dense.
Stats out today show net migration flows continuing to swing strongly positive.
= Net gain of 15,000 in the year to Sept., compared to minus 3,200 a year earlier.
And 2,700 (seasonally adjusted) last month alone. (Actual number 4,194)
It is not unconceivable that we will see annual inflows well in excess of 20 or 25 thousand, as N.Z. and Australian economies diverge over the next year or two.