Impact of COVID-19 on international migration

Since January 2020, governments have imposed international travel restrictions in multiple countries, due to the spread of COVID-19 around the world. In March 2020, the New Zealand government introduced further measures to protect New Zealanders from the COVID-19 virus, effectively limiting travel to New Zealand and travel within New Zealand.

The comparisons between the September 2019 and 2020 months and years use the latest provisional estimates for each period.

The provisional estimates have 95 percent confidence intervals (±) beside them – the wider the interval, the greater the uncertainty about the estimate. However, these intervals reflect the model uncertainty, not the extent of future revisions to provisional data.

Provisional migration estimates are revised each month until they are finalised after 16 months. Migration data transformation has more information about the migration estimates, including the 12/16-month rule.


Year ended September 2020 (compared with year ended September 2019) provisional estimates were:

  • migrant arrivals – 130,100 (± 1,400), down 12.1 percent
  • migrant departures – 62,300 (± 600), down 29.7 percent
  • annual net migration gain – 67,700 (± 1,500), up from 59,400 (± 200).

For migrant arrivals in the September 2020 year, New Zealand citizens were the largest group with 44,000 (± 600) arrivals. The next largest groups were citizens of:

  • India – 13,200 (± 300)
  • China – 8,500 (± 200)
  • South Africa – 7,100 (± 100)
  • United Kingdom – 6,600 (± 100)
  • Australia – 5,900 (± 200)
  • Philippines – 5,000 (± 100).

For migrant departures in the September 2020 year, New Zealand citizens were the largest group with 20,300 (± 300) departures. The next largest groups were citizens of:

  • China – 10,700 (± 300)
  • India – 4,900 (± 100)
  • United Kingdom – 3,500 (± 100)
  • Australia – 2,800 (± 100).


September 2020 month (compared with September 2019 month) provisional estimates were:

  • migrant arrivals – 2,900 (± 200), down 80.7 percent
  • migrant departures – 2,100 (± 200), down 69.2 percent
  • net gain – 800 (± 200) migrants.

Implications of worldwide travel restrictions on recent migration estimates

  • Many visitors to New Zealand, travelling on a range of visa types, have been unable to depart (estimates as at 6 November 2020 are 80,000–110,000 visitors in New Zealand; see also 25 March 2020 discussion in About 250,000 visitors in New Zealand). Migration estimates could be revised up or down depending on whether these people stay in New Zealand or head back overseas. Their prolonged stay in New Zealand is currently contributing to high estimates of migrant arrivals and net migration between late-2019 and March 2020.
  • The travel of New Zealand residents has been curtailed, resulting in relatively fewer short-term and long-term (migrant) departures of New Zealand citizens in recent months relative to the same period a year ago.
  • In addition, estimated migrant arrivals of New Zealand citizens was higher than usual up to March 2020. If these people subsequently depart rather than staying long-term, then provisional migrant arrival estimates for these months are likely to be revised down.

Migration Data Transformation project – Comparison of provisional and final migration estimates: August 2017–September 2020 has a monthly summary of the number of border movements, the number requiring modelling as migrants or non-migrants, and the history of revisions to migration estimates.

Revision to historical migration estimates

As outlined in previous releases, we have been improving the data used by the estimation system that deterministically classifies travellers as migrants or non-migrants and identifies which traveller records require modelling of migrant status.

With this release of provisional migration estimates for September 2020 we have revised final migration estimates from January 2001 to May 2019. This includes minor revisions to all months, quarters, and year ended periods in all published outcomes-based migration series, including seasonally adjusted and trend series.

These revisions incorporate an improved approach to identifying unique travellers through the use of extra administrative data in the Integrated Data Infrastructure (IDI). This enables us to construct more accurate travel histories leading to improved classifications of border-crossings as migrant or non-migrant.

In October 2019 with the release of August 2019 data, we released provisional migration estimates based on improved travel histories. At the time extra administrative data was not available to extend the improved travel histories for the entire outcomes-based migration back series. However, we now have the data to more accurately calculate the migrant statuses of all travellers across the entire series.

The largest monthly revision occurs for departures in May 2013 (down 700, or 8 percent) and arrivals in June 2013 (down 1,031, or 12 percent).

  • When the outcomes-based migration series was first published in January 2019, linked travel histories only went back to June 2013. Because of this, travellers’ migrant status was classified from self-reported passenger card information. Extending travel histories back further has allowed us to calculate migrant status more accurately, causing the relatively larger changes in May 2013 and June 2013 migration estimates, with flow-on effects through subsequent months.

In addition, the size of revisions in monthly migration estimates differ between the periods of January 2001 to April 2013, and July 2013 to May 2019, respectively.

  • Arrivals in finalised months from January 2001 to April 2013 have been revised down by an average of 14 (0.2 percent), and departures revised down by an average of 20 (0.3 percent).
  • Arrivals in finalised months from July 2013 to May 2019 have been revised down by an average of 172 (1.5 percent), and departures revised down by an average of 216 (3.0 percent).

As arrivals and departures have typically revised in the same direction, the effect on net migration has been low, with the majority of monthly estimates revising by no more than ±150.

Other statistics that use migration data as an input are not impacted by this change as revised estimates were incorporated internally at Stats NZ.

This update also resolves a previously advised inconsistency where year-ended migration estimates from January 2018 were inconsistent with the sum of the monthly estimates.

Why migration estimates change

Before November 2018, we classified travellers based on the intentions they stated on their passenger cards when they crossed the border. Despite the uncertainty inherent in provisional estimates, calculating actual travel duration (outcomes-based measure) is a more accurate way of measuring migration than relying on passenger card intentions. The outcomes-based measure is available back to January 2001.

From November 2018, we implemented the new outcomes-based approach, which uses passport data to link arrivals and departures and accurately measure how long people spend in, or out of, New Zealand after their initial border crossing. To classify a border crossing as a migrant movement, we need to observe up to 16 months of travel history, using the 12/16-month rule.

With this new approach, it takes 17½ months before final migration estimates are available. To produce timely results, we use a statistical model to produce provisional migration estimates. Statistics produced using these provisional estimates have uncertainty for 16 months; after which time, we can finalise the classification of all border crossings.

As new data becomes available, the provisional migration model has more information about the border crossings it is trying to estimate. So, with an extra month of data available, this causes shifts in the estimated number of migrant arrivals and migrant departures and thus changes in the net migration estimates.

For example, the extra data will indicate travellers who have now departed New Zealand, or travellers who were away and have since returned to New Zealand.

In contrast to the international migration estimates, international travel statistics are still based on the stated intentions of travellers when they cross the border. For example, the classification of travellers as ‘NZ-resident travellers’ or ‘overseas visitors’ is still largely based on their responses on the arrival card. If an overseas visitor ends up staying for 12 months or more, they will be counted as a migrant arrival, but overseas visitor statistics are not revised to reflect their eventual status. Any revisions to short-term travel statistics would generally be small relative to the large volume of short-term travellers.

Early and provisional international travel data

Each week, Stats NZ  releases early and Provisional international travel statistics to facilitate analysis of the COVID-19 international pandemic and the impact on New Zealand’s inbound and outbound tourism sectors. This includes:

  • weekly travel data for short-term overseas visitor and New Zealand resident arrivals
  • arrivals and departures data based solely on border-crossings into and out of New Zealand
  • stock estimates of visitors in New Zealand, and New Zealand residents travelling overseas, based on arrivals and departures.

Leave a Reply