Crystal ball gazing – predictions for the 2018 New Zealand Census results

Last week Penny and I attended the SOLGM conference ( New Zealand Society of Local Government Managers), where I presented my 12 predictions for the results we expect to see after the New Zealand Census data are released next March.

We have been working with New Zealand Local Government to help understand Census data and forecast populations for the past 8 years, and have been major sponsors of the SOLGM planning forum for 7 years.

Census is still our primary source of demographic information for planning at the very small area level. Our online tools present this in a way to help you get on with telling the story of your area to inform your planning decisions. Our forecasts give you a strong evidence base for understanding your future population on the basis of an agreed set of assumptions.

We were a bit concerned to hear that the NZ Census in 2018 has not gone entirely smoothly – reports are that up to 10% of the population has been missed.

Australia had similar issues with our 2016 Census – the website going down on Census night and a lower response rate, nevertheless up to 96% in the end. We have confidence that Stats NZ will be able to address the missing data and provide a robust and usable Census, but there will likely be some limitations, which we’ll watch closely.

In the 2011 and 2016 Censuses, I made some predictions about what the results would show for Australia’s rapidly changing population. You can see how I did here.

Our work with Local Government across Austraila and NZ gives us some great insights into what each region is dealing with in terms of population, so I put together a series of predictions – exact numbers and percentages. It’s essentailly crystal ball gazing, but with some evidence behind it – about what the Census may tell us about the community.

So for 2018, I’m doing the same for New Zealand. For 2016, I got 7 hits of 12 predictions in Australia, plus two partial hits, with the direction but not the size of the change.

So here are my predictions for the New Zealand 2018 Census results

1. Revised Population estimate for New Zealand. Currently 4,871,300 for March 31. Census estimate will bring the population up slightly, to 4,885,200. The Census count for New Zealand will be 4,560,290.
2. More people missed. Not so much as a prediction as a fact now – Stats NZ are on record saying that they think only 90% of people responded to the Census, though most did so online. This means more imputation – missed households inherit the characteristics of similar households nearby. The total percentage of “Non classifiable” households will more than double, from 2.6% in 2013, to 6.0% in 2018.
3. Regional population growth. Of the growth of 318,248 people between the 2013 and 2018 Censuses, 249,000, or 78% will be on the North Island, and 185,000, (58%) will be in Auckland. 80% of the South Island’s growth will be in the Canterbury region, which will add 50,000. However, there will be no New Zealand region which records negative growth – though there may be at the TA level.
4. Age structure. The median age will not change from 38 years, recorded in 2013. This will be the first Census since the 1970s that the population didn’t get any older, on average. This is mainly due to increased migration of younger people, offsetting the ageing of those already here.
5. Changing migration. Net permanent and long-term migration to New Zealand has changed enormously since 2013, and is now running at 70,000 per annum. Expect to see a tick up in the number of Australian-born residents, from 62,000 to 75,000. But the big increases will remain people from China (up by 40,000 to 130,000), India (up 35,000 to 102,000) and the Philippines (up by 20,000 to 60,000 in total). Smaller, emerging countries like Burma (Myanmar) and Nepal and Saudi Arabia will continue to feature.
6. Changing languages. Maori will remain the most widely spoken language after English, but the numbers will continue to slip, down to 135,000 people in 2018. Chinese languages will number 130,000 when combined. There will be a greater undercount for those speaking a non-English language, due to the issues with the Census.
7. Changes in religion. New Zealand has one of the highest rates of “No Religion” anywhere in the world, with 38.6% selecting this option in 2013. In 2018, Stats NZ have changed the religion question from tick-box to write-in and moved “No Religion” to the bottom of the question. Because of this the growth in ‘No Religion’ will not be as large as previously, increasing only slightly, to 40.5% of population. Former tick-box categories for the major Christian religions such as ‘Catholic’ and ‘Anglican’ will fall, and ‘Not Stated’ will increase, bringing total Christian religions down to 40.2%, just below the ‘No Religion’ response.
8. Rentals rise, home ownership falls. Large housing price increases will see an increase in the proportion of households rented nationwide. Renting will take over as the #1 tenure type, at 30.5% of all households, while mortgages fall below 30%. Nowhere will this be more evident than Wellington City Council, which will top 40% renters for the first time.
9. Housing for young adults. 25-34 year olds will be particularly affected by housing affordability. In 2013, the proportion of this group living in a dwelling they did not own was 68%. In 2018, this will rise to 74%.
10. New Zealand’s largest industry. Health Care and Social Assistance became New Zealand’s largest industry in 2013, and this will not change. Health Care will hit 225,000 people employed, just over 10% of all workers. Manufacturing and Agriculture will employ less people than ever.
11. Higher education. We continue to become more educated. The proportion of New Zealanders with a Bachelor Degree qualification will rise from 18% in 2013 to 22% in 2018.
12. Stable and slightly increasing household size. The average household size in 2013 was 2.66 people per dwelling. With the housing affordability squeeze and increasing migration, this will turn around, reversing earlier declines, going back up to 2.70 people per dwelling.

So there you have it – 12 predictions for the 2018 Census results!

The best estimate is that these will be due for release in March 2019. We will get them into our community profiles as soon as possible, and they will become the basis for our new population forecasts as well.

And, when the results are released, I’ll write another blog to look at how well (or badly) I did!