Experimental estimates of income from linked administrative data

Project Description

 Give feedback

This series of experimental income estimates has been derived from the tax data available in the IDI as part of ongoing work at Stats NZ to increase the use of administrative data in the production of statistics.

You can view the following:

We are publishing these estimates to invite feedback to support future development of income measures from administrative sources.

These experimental income estimates are NOT official statistics. Rather, they are published as an example of the type and quality of information about income that can currently be obtained from administrative data sources.

Stats NZ continues to publish official statistics about income. These are available from the Income page on the Stats NZ website. The User guide for Statistics New Zealand’s wage and income measures can help you identify the most appropriate measure for your needs.

What you'll see

You will see a series of income estimates at national, regional council and territorial authority area level by five-year age groups and sex for the tax years 2010 to 2016. For each year, median income is provided for all groups and for each geographical area the distribution of income is available. Experimental estimates of income from linked administrative data: Methods and results is a technical paper that describes the methods and data sources used to produce the experimental estimates.


We'd like your feedback about:

  • the overall usefulness of
    • the series estimates
    • the accompanying technical report
  • any suggestions for possible extensions to the scope of the estimates, and how you would like to see these prioritised if the series is further developed in the future
  • anything else you would like to see in future releases.

To give feedback, click on the 'Give feedback' button above.

Interactive app

In order to make it easier to view the data you want, we have provided an application below that lets you select and download specific breakdowns of the income estimates. In the grey box you can select the variables you wish to view or download. You can click on the selection box and add any number of variables. To remove a variable, click on it, then use the delete or backspace key. The download button will give you a CSV file of the selected data.

‘Count’ is the number of individuals aged 15 and over with income information available on which the median is based

‘Median’ is the median annual personal income from specified sources in NZ $

S = Suppressed to protect confidentiality


Data source

This series has been produced using administrative tax data in the IDI and the experimental population estimates (IDI-ERP).

Data quality

This series of experimental income estimates derived from administrative data has some benefits compared to existing official estimates. However, there are also some known limitations that users should be aware of.

Potential benefits include increased accuracy from using administrative tax data rather than relying on respondent recall and disclosure of sensitive information to an interviewer, and the detailed geographic breakdown that using linked administrative sources (IDI) allows.

The main limitations relate to any error in the target population which will affect the accuracy of the income estimates accordingly (See Experimental population estimates from linked administrative data: 2017 release for more details) and the timeliness with which information about income from self-employment is available due to the annual tax return cycle. Benefits and limitations of the estimates are discussed more fully in Experimental estimates of income from linked administrative data: Methods and results.

Known issues


Release notes

Version 1. This is the first set of experimental income estimates from the IDI.

Published 6 October 2017

Project Details