This page explains the updates to the data sources used to compile the international trade release. It also explains two corrections made to processing that affects several services series. This data will be published as part of the international trade release on 3 September 2024.
We make data updates annually, in line with international best practice, to ensure that our statistics continue to reflect our changing world.
The changes made in this release fall into two categories:
- new or updated information becoming available - this covers data from the September 2018 quarter to the June 2024 quarter
- corrections to our outputs due to internal errors in data processing - these cover data from the September 2022 quarter to the March 2024 quarter.
Contents
Travel services exports
Education travel services exports
Education travel services imports
Insurance services exports and imports
Personal, cultural, and recreational services exports
Other data updates
Travel services exports
During Stats NZ's March 2024 quarter releases, the travel services exports series results had an unexpected effect on the corresponding seasonally adjusted series in other releases. Seasonally adjusted estimates of travel services exports in the March 2024 quarter were unexpectedly low. We identified that this was the result of differences in the seasonal pattern produced by the pre-2020 and post-2020 methodologies. We acknowledged this issue in our published data in the March 2024 quarterly releases.
Macroeconomic statistics affected by this series include balance of payments (BOP) statistics, household consumption and expenditure (HCE) and household saving. There will also be a minor impact on aggregate measures of gross domestic product (GDP).
Interim solution
Following investigation, we will implement an interim solution to improve the consistency of the travel services exports series. We determined that to produce robust seasonally adjusted estimates, it would be necessary to use consistent pre-2020 and post-2020 methodology.
Our solution is to return to the direct use of estimates from the International Visitor Survey (IVS) from September 2022 onwards to be consistent with the methodology of the pre-2020 series. We will retain the existing modelled values through the COVID-19 border closure period (March 2020 to June 2022) which were based on the Visitor Stock Model (VSM) when the IVS was not able to be run.
Travel services exports models - background information has further information on the difference in methods and the timing issue it caused.
Stats NZ will revise quarterly travel services exports figures from September 2022 to March 2024. This revision will improve the data quality of the corresponding seasonally adjusted series and the macroeconomic statistics that rely on it.
The revised figures will be included in the International trade: June 2024 quarter release on 3 September. Note that education-related travel exports are being revised separately and are not included in the impacted estimates in this update.
Impact on travel services exports
The largest impact on total travel services exports estimates is in the March 2024 quarter which will be revised up by approximately $900 million. For all other quarters from September 2022 to December 2023, revision size ranges from decreases of $300 million to increases of around $100 million.
In the March 2024 quarter, the revision size is an increase of approximately $900 million to other/personal and business travel services. This will lead to a downward revision of household consumption expenditure (HCE) and an upward revision of the current account balance and household savings.
Impacts on other macroeconomic statistics
Balance of payments
The updates to the travel services exports values will affect balance of payments statistics through a change to the current account balance (CAB). An upward revision to travel services exports would reduce the current account deficit in a given quarter, whereas a downward revision would increase the deficit.
HCE and household saving
An upwards revision to the travel credits estimate will contribute towards a downwards revision of HCE estimates. Visitor spending in New Zealand is subtracted from total spending in New Zealand during the calculation of HCE.
Household saving estimates are derived by subtracting HCE from household disposable income estimates. This means that an upwards revision in travel credits will cause an upwards revision in household saving estimates.
GDP
GDP is not directly affected by this revision to travel credits. Our HCE estimates flow into unallocated taxes and therefore travel credits indirectly affect the final GDP calculation. We anticipate the effect on unallocated taxes and GDP will be small.
Within the expenditure measure of GDP, impacts on HCE and exports will largely cancel out. There will, however, remain a small impact on the expenditure measure of GDP as business travel credits are included within exports but are excluded from HCE. There will be further small impacts on the expenditure measure of GDP due to the differences in chaining weights within exports and HCE.
Travel services exports models - background information
Until March 2020, travel services exports (international visitor expenditure) in New Zealand was measured through the International Visitor Survey (IVS).
The IVS was not run between March 2020 and September 2022 due to COVID-19 border closures. In 2020, the new visitor stock model (VSM) was developed to maintain measurement of this important macroeconomic statistic.
The VSM uses arrival and departure data to estimate a 'stock' of visitors in New Zealand during a given quarter, and has been in use since the March 2020 quarter.
Estimating visitor expenditure in New Zealand during the June 2020 quarter has more information about the introduction of the VSM for travel credits in 2020.
The current visitor stock model (VSM)
The VSM relies on two data sources:
- arrivals and departures data from the New Zealand Customs Service
- reported spending by travellers while in New Zealand from the IVS.
Stats NZ estimates a 'stock' of the number of non-resident travellers in New Zealand on a given day, which gives us the number of person-days in a quarter.
Total international visitor expenditure is then calculated by multiplying the number of person-days by the average spend. Average spend and person-days are broken down by country, length of stay, and purpose of travel.
Timing differences between IVS estimates and the VSM
The advantage of using a visitor 'stock' in this model is the timing accuracy that can be achieved from the Customs data. The VSM allocates the spending of visitors to when they are in New Zealand, whereas the IVS estimate allocates the spending of visitors to the date they depart New Zealand. For visitors whose stay straddles two quarters, some of the spending is measured in different quarters in the IVS and the VSM. This difference is a timing inaccuracy in the IVS estimates.
The switch from the IVS estimates to the VSM in 2020 introduced differences in seasonal patterns due to this slight difference in measurement timings between the two methods.
Travel exports have now returned to a similar seasonal pattern to those seen before 2020. This has revealed an effect of the timing difference between these two methods, seen in the seasonally adjusted series for March 2024.
Review of travel services methodology
We are reviewing and improving the VSM and it is not yet ready for use across a longer time series. While we are working on improving and backdating the VSM, we will revert to using IVS estimates to reduce inconsistency between pre- and post-COVID-19 time series.
In the long term, we expect adopting the VSM will improve the quality of estimates of travel services exports. This model may be able to be augmented or fully converted by using more administrative data sources, such as electronic card spend data. This approach may also be used to improve travel debits estimates.
Education travel services exports - expenditure by international students in New Zealand
We have revised the March 2023 to March 2024 quarters (inclusive) to be based on the latest Export Education Levy data compiled by the Ministry of Education and updated student visa numbers compiled by Immigration New Zealand.
The largest impact on education-related travel services exports estimates is in the March 2024 quarter, which will be revised upwards by approximately $200 million due to timing issues with irregular data input - which led to a result not representative of wider trend.
Revisions to other quarters are all downwards revisions, the largest of which is -$15 million.
Education travel services imports - expenditure by New Zealanders studying overseas
We have revised the June 2023 to March 2024 quarters (inclusive) to better reflect prices of tuition and living costs for New Zealanders studying overseas, and to use updated New Zealand resident arrival numbers.
Insurance services exports and imports
We have revised the March 2021 to March 2023 quarters (inclusive) to be based on updated source data and validation of the series against other information sources, such as annual financial statements of companies.
Personal, cultural, and recreational services exports
We have revised film exports using external data sources. This affects the September 2023 to March 2024 quarters (inclusive).
Other data updates
We have made minor updates to other series in the September 2018 to March 2024 quarters (inclusive), based on updated data provided by individual enterprise survey respondents.
Technical enquiries
Henry Han
04 931 4600
[email protected]
