Net overseas migration: why is it so high?

Author:Birrell, Bob

Between 2004-05 and 2007-08 net overseas migration (NOM) increased dramatically. In 2006 the ABS introduced a new methodology to measure NOM. This article shows that the effect of this new methodology on NOM estimates was minor. Most of the growth was real and was mainly attributable to increases in the temporary entry visa categories, especially overseas students. The article also shows that the net loss of citizens and permanent residents--the alleged brain drain--is much smaller than previously assumed.


According to the Australian Bureau of Statistics (ABS), net overseas migration (NOM) increased from 123,800 in 2004-05 to 277,330 in 2007-08. These are final estimates. The preliminary estimate for 2008-09 is 299,000. The figures quoted above incorporate a change in the methodology used to estimate NOM from the September quarter of 2006. There were two major changes in this new methodology. The first was a new measure of who is included as a resident. A resident is anyone who has stayed in Australia for 12 months out of 16 months. It does not matter what their residence status is. To be counted as a departure, a resident must have left Australia for at least 12 months out of 16 months. (Henceforth, the new measure is referred to as 12/16.) Prior to this change, a resident was regarded as one who had stayed in Australia for a continuous 12 months and a departure was one who stayed away for a continuous 12 months.

The second change was that the ABS used a traveller, rather than a movement, database in calculating who was in or out of Australia according to the above definition. This aspect of the methodology is explained below. The striking increases in NOM since 2004-05 have attracted considerable attention and controversy. In part, the controversy relates to the methodological changes just described for estimating NOM. Some commentators have argued that the increase in NOM may be partly an artefact of these changes and that, for this reason, concerns about the scale of the increase have been unnecessarily alarmist.

In this paper, we first discuss the way the ABS measures NOM and the implications for estimates of NOM flowing from the changes to the methodology introduced in 2006. The paper then examines the growth in NOM between 2004-05 and 2007-08 by the visa category and the major country of birth of those counted as arrivals and departures during these years. The purpose is to get a better understanding of the source of growth in NOM. As discussed below, an accurate understanding of these sources is important because immigration planning and estimates of future population are in part based on assumptions about the scale of NOM.

This analysis uses a customised set of NOM data by visa class for the years 2004-05 to 2007 08. Final data for 2008-09 must wait until records of actual movements of travellers become available for the 16 months after arrival or departure in 2008-09. The methodology used by the ABS for this data set is that introduced in 2006 (that is, it incorporates the 12/16 rule). Thus, the estimates for 2004-05 and 2005-06 are comparable with those for 2006-07 and 2007-08.


Under the new traveller-based methodology, the ABS compiles information on the movements of all travellers who arrive in or leave Australia each year, whether they be visitors, students, temporary workers, those holding permanent resident visas, New Zealand citizens or Australian citizens. This is based partly on data from the passenger cards that travellers complete when leaving or arriving in Australia and partly on administrative records obtained by the Department of Immigration and Citizenship (DIAC) for its Travel and Immigration Processing System (TRIPS). From these records the ABS is able to calculate and estimate NOM each quarter. It does so by using a unique identifier for each traveller in order to construct a traveller history for all arrivals and departures.

The new methodology has two benefits. The first is that it removes inaccuracies deriving from persons who arrive or depart on multiple occasions. The creation of the movement record for each traveller removes a potential source of double or multiple counting. Previously, such movements had to be estimated by matching incoming and outgoing passenger cards. (1) The second benefit is that the new methodology permits the calculation of the actual stay in or away from Australia by each traveller and can therefore be used to count those who meet the 12/16 measurement rule for the purposes of calculating NOM. The ABS can now measure the actual behaviour of travellers in terms of the duration of their stay or departure. Under the new methodology, NOM is the difference between arrivals who stay for 12 months or more over 16 months (who are added to the population) and residents who stay away for 12 months or more over a 16 month period (who are subtracted from the population). (2)


According to ANU demographer, Peter McDonald, the ABS did not count movements of temporary residents prior to the introduction of the new methodology in 2006. He states that: 'since 2006 people who enter Australia on a long-term temporary basis have been counted as migrants. If they had been counted before this, migration during earlier years would have been higher'. (3) This view has been widely disseminated. However, it is not true. The ABS has always counted temporary residents, regardless of the purpose of their stay, if they meet the prevailing definition of a resident for NOM purposes. As noted, the measurement of the time of stay or departure changed in 2006, but not the range of travellers eligible to be counted as part of NOM.

This change is potentially important. As the current Minister of Sustainable Population, Tony Burke, has said, the old methodology 'knocked out most overseas students' because a very large number visit their families one or more times each year. (4) Burke does not imply that the new methodology is ill-advised, but rather that it may have had a large impact on the NOM estimates. In our view, it is appropriate to include people as residents if they stay for a minimum of 12 out of 16 months. To suggest that overseas students are not part of Australia's population because many go home once or twice each year is misleading. Their presence in the major metropolises is palpable and their demand for housing, transport and other services has a major impact on these cities.

Table 1 provides a...

To continue reading