Kudos to the Australian Taxation Office, who have just released a 1% sample of Australian taxpayers for the use of researchers.

One of the reasons that empirical public finance has been comparatively weak in Australia is the paucity of good data. This dataset will do more to change that than millions of dollars in ARC grants would have done. Here’s the blurb:

This year the ATO embarked on an ambitious project to produce and release a confidentialised 1% sample file containing individual tax return information. This was to satisfy the requirements of more advanced users of taxation data and further demonstrates the ATO’s commitment to an open and transparent tax system.

The file is confidentialised in order to protect the identities of those contained in the file. This year we are releasing two sample files. A file for the 2005–06 income year based on data contained in Taxation statistics 2005–06 and a file for the 2006–07 income year based on data in Taxation statistics 2006–07. Each year the file will be released in conjunction with the release of Taxation statistics.

The file may be accessed by anyone conducting legitimate research. An application for accessing the file can be obtained by emailing taxstats@@ato.gov.au.

For more details on the contents, see the ATO website. My hope is that once this has been running for a few years, the ATO will also release a panel. A cross-section is useful, but there is much more that can be done when you can follow the same individuals over time.

To any honours or PhD students currently choosing a topic: I’d highly recommend using this dataset. There are many important Australian tax questions that can now be answered much better than ever before.

(Crossposted at andrewleigh.com)