The ABS recently released a new research paper titled, “Feasibility Study of Linking Migrant Settlement Records to Personal Income Tax Data“, authored by Laura Walsh and Anne Weckert (h/t Gaby D’Souza).
The project aims to match data from ATO income tax records and the Settlement Database, which tracks (some) migrants. This will build on Australia’s already excellent migration statistic framework.
Most of the paper is methodological in nature. I have read it and have nothing more to say given I’m completely ignorant about complex statistical methodologies.
However, if the methodology holds up (which I assume it will), the potential for new research is profound. Here are two examples from the research paper:
Given the preliminary nature of this research agenda, these results “are uncalibrated and are considered experimental”.
This type of information is important. We know, as per the first graph, that migrants on average increase their salaries over time in Australia. We also know as per the second graph, that partners and spouses on average earn less than primary visa holders.
But this level of detail and precision can assist in broader comparisons about the labour market, particularly with native Australians (native is the common term in labour economics to describe those born in the country of reference). We can see for instance how different cohorts of migrants adjust and over time periods.
This becomes especially helpful for immigration policy-makers who can match the results from this research agenda against visa policy settings for different time periods. There are of course a range of other potential benefits to this exercise, many of which will mean we can better understand how Australian migration policy impacts our society and economy.
One note of caution. From my experience at Immigration, the Settlement Database (being used by the ABS to match records) was never the tool I thought it was. There were often significant gaps from missing or dated information. Processes around how the information was managed were also lacking (to a comical degree). This was now sometime ago and I believe progress on these issues has likely been made. However this reminds me that sources of information can be overlooked in terms of how much validity we can place on records.
Lastly, a quick comment on the ABS. This is where the ABS is a pioneer. Survey work is their bread and butter and something they do extremely well. But this type of work highlights the massive potential the ABS has to inform a wide range of policy areas. This is a strong part of the reason why a well funded ABS is so critical across government. Cutbacks of staffing and research capacity might save money in the short-term. But this is an excellent example of where a small team working on an important long-term project will contribute to the better understanding of how we make immigration policy with serious future benefit for all Australians. Trying to sort this out in a cost/benefit model via efficiency dividends is impossible.