How many NZers? Economists say 'who knows'
Wednesday, 12 June 2019
Economists have been left perplexed by new migration data that is meant to provide a clearer picture of what's really happening.
Since February, migration statistics from Stats NZ has been 'outcomes based', collated using passport data and arrivals cards.
It replaces the old 'intentions-based' system which relied on what passengers said on their now-defunct departure cards.
But people cannot be classified as having moved here permanently until enough time has passed with no indication they have left.
**READ MORE:
* Departure cards to be gone by Christmas
* More migrants leaving NZ, more UK and Philippines citizens want work visas
* New Government's policies could hurt housing market**
In February, it was reported that immigration was much lower than expected. Then, in March, Stats NZ said the numbers ticked up again. This week, it said annual net migration had been at high levels since 2014.
Migrant arrivals were provisionally estimated at 151,000 and migrant departures at 95,100 in the 12 months ended April 2019, the most recent data available. This resulted in a provisional estimate of annual net migration of 55,800.
Record levels of migration have been a focus over recent years, as the main driver of a booming population putting pressure on infrastructure and house prices in some parts of the country.
But question marks over the data have prompted economists Infometrics to decide to stop covering net migration in monthly news releases.
'We believe the volatility in the most recent months of StatsNZ's estimates makes this release of little value to you, our clients. However, we are aware StatsNZ is continuing to refine and improve its model, and we will keep an eye on their output to decide if it is worthwhile restarting this release,' they said in a newsletter.
'The continual downward revisions lead us to believe StatsNZ's model is overcooking its figures, particularly in the last seven months of their estimates.'
Infometrics chief forecaster Gareth Kiernan said there were issues to work through for Stats NZ.
'They only have good data back to 2013/14 on which to model traveller behaviour for the 'unknowns' over the last 17 months. Because net migration was mostly climbing between 2013 and 2017, the rising trend might be biasing their model estimates for the latest months upwards.
'I think the reality is that permanent and long-term migration is such a small component of overall passenger movements, so there is always going to be a high degree of variability in their estimates for the most recent six months or so.'
He said it would have been good to run the two migration measures together for a time before the departure cards were phased out.
'The political rush to get rid of the cards and save money has meant that approach was not possible.'
ASB chief economist Nick Tuffley said the new data was a lot less useful because it was subject to so much potential revision.
It would take 16 months before it was concrete enough not to require estimates.
Stats NZ classifies a migrant arrival as an overseas resident who arrives in New Zealand and spends at least 12 of the next 16 months here.
A departure is a New Zealand resident who leaves and spends at least 12 of the next 16 months out of New Zealand.
'It's not as certain as it used to be how fast the population is growing.'
It had been compounded by problems with the Census, he said, and Stats NZ's decision to pull the accommodation survey.
Stats NZ senior demographer Kim Dunstan said the new statistics gave a clearer picture than was possible to achieve with intentions-based measures.
'There's no doubt New Zealand net migration remains at high levels.'
He said even though the data was subject to revision it was not going to change by tens of thousands of people.
Provisional estimates would become more stable with each month.
Focus was being put on November 2018 now, for which Stats NZ said its figures were now reliable.
'But that doesn't mean that any customers, including bank economists, have to wait for a final estimate to be useful.'
He said the previous intentions-based data had understated both the number of arrivals and departures.