INTERSTATE MIGRATION

9.77 Interstate migration is a key determinant of the accuracy of state and territory population estimates. Data on interstate migration cannot be directly estimated unlike that of natural increase and net overseas migration. Instead, post-censal quarterly estimates of interstate migration are modelled using administrative by-product data. Over time, the ABS has used a number of administrative data sources to produce quarterly estimates of interstate migration, including electoral roll registrations and family allowance payments. Currently the data used by the ABS is information on interstate change of address advised to Medicare Australia and to the Department of Defence in the case of the military.

9.78 The Medicare-based model used for generating post-censal estimates of interstate migration is largely superseded when new Census information becomes available. For example, every five years, after data from the following Census has been finalised, the modelled estimates are reviewed against, and potentially replaced by, the interstate migration estimates that are calculated from the Census (i.e. rebased to the Census). This is known as the re-derivation of interstate migration and is mentioned in Chapter 7 - Estimating interstate migration.

9.79 Due to incomplete coverage and the non-compulsory nature of available administrative (indirect) data sources, post-censal quarterly estimates of interstate migration have long been considered the weakest measure of a component of population change.


Criteria for data source selection

9.80 In Australia there exist a range of administrative data sources that might be considered for modelling interstate movements. These include Medicare data, electoral roll enrolments, social security, taxation data and drivers licence information. However, to be of use these data sources must meet a number of criteria. These include:

• population and coverage - the source must provide complete, or near complete, coverage of the usually resident population of Australia
• data content - individual data records must include certain data items (such as age/date of birth, sex, date of move, origin and destination)
• geographic level - data must be available at the state/territory level
• timeliness - data must be available shortly after the end of the reference period
• historical availability - to be able to model, a relatively long time-series of data is required
• consistency - there must be no or minimal definitional changes to the data over time (or at a minimum, changes must be quantifiable)
• electronic capture - current and past data must be available in electronic format.

Strength of Medicare data as an indicator

9.81 The ABS evaluates the range of potential sources of administrative data for its effectiveness in estimating quarterly interstate migration. Medicare Australia data supplying change of address information was found to be the most effective source currently available. For more information, refer to the Information Paper: Evaluation of Administrative Data Sources for Use in Quarterly Estimation of Interstate Migration, 2006 to 2011 (cat. no. 3127.0.55.001).

9.82 Medicare card holders are required to register changes of address when they make claims, or when their cards are replaced. However, it is known that some people, particularly younger card holders, do not register changes of address with Medicare, or do so long after the fact. Comparison of Medicare change of address with Census data on the address of respondents one year prior to the Census suggests that the level of undercoverage in Medicare is fairly constant over time. In addition, this under-reporting seems to be similar for interstate arrivals and interstate departures, as well as for each sex by various ages.

9.83 The undercoverage of interstate movers within Medicare data is adjusted by the ABS for each sex at specific ages through the use of expansion factors. These expansion factors scale up Medicare data for these specific ages to reflect the higher levels measured by information provided by the most recent Census. For the intercensal period 2006 to 2011 expansion factors are to be applied to males aged 17-30 years and females aged 17-25 year universally for each state and territory for both arrivals and departures. For more information see Chapter 7 - Estimating interstate migration.

9.84 Medicare theoretically covers all Australian usual residents as well as those non-Australian residents granted temporary registration. However, there are a range of Australian usual residents who do not access the Medicare system, primarily due to access to alternative health services, such as defence force personnel, prisoners, persons eligible for the Department of Veterans Affairs Health Services and Indigenous Australians. Strategies have been put in place however, to enrol Indigenous Australians with Medicare Australia who would normally just use an Aboriginal medical service.


Defence force adjustment

9.85 Adjustments to compensate for interstate defence force movements not covered by Medicare are applied to the quarterly interstate migration estimates. These adjustments are estimated using counts of defence force personnel by age, sex and state/territory, obtained from the Department of Defence, with 70% of any change in quarterly defence force numbers assumed to be due to interstate migration not otherwise covered by the Medicare model.


Changes to source data over time

9.86 The first attempt to compile interstate migration estimates was made for a period prior to 1966 when interstate arrivals and departures were estimated from records of all movements (including short-term movements) by sea, air and rail. From June 1966 to 1981, interstate movements were based on records of interstate changes of address on child endowment (family allowance) registers and electoral rolls. Holiday, business and other short-term movements were no longer taken into account. The change was made in the recognition that estimates based on air, sea and rail travel were inadequate as measures of total net interstate movements (travel by car, for example, was not recorded) and that measuring all movements including short-term moves was neither necessary nor practicable.

9.87 At the time, the family allowance was paid universally (to all mothers of children under the age of 17 years) but in November 1987 means testing was introduced. An upper income limit for entitlement was set, graduated according to the number of children in the family.

9.88 The Department of Social Security (DSS) was unable to estimate the impact of means testing. In October 1987 there were 2.1 million families collecting the allowance, one month later it declined to 1.8 million. It then started to gradually increase, reaching 1.9 million in August 1989. At the time of introduction of means testing the Department had no data on income of parents. As a result, the department could not exactly explain whether the decline since November 1987 was due to:

• income testing; or
• families with the youngest child turning 17 years (and therefore becoming ineligible).

9.89 Whatever the answer, it was clear that means testing, by reducing coverage of the population, reduced the effectiveness of the family allowance register as an indicator of interstate migration. In addition, effectiveness was subject to changes over the years to the age limit of children for which family allowance was applicable.

9.90 Electoral roll data were used in conjunction with family allowance data, but were much less reliable as a source of interstate migration statistics. While the registration of electors, voting, and notification of a change of address are compulsory in law, several factors affect the quality of the data. First, a proportion of the population are not entitled to vote and are therefore not on the electoral roll. Second, registration of persons entitled to vote is incomplete. Third, the timing of elections has affected the timing of notification of change of address. Fourth, the timing of Electoral Office surveys to check the coverage of their rolls (called 'habitation' reviews) also affect the timing of notification of change of address.

9.91 All these factors affected both the coverage of the data and the reliability of the interstate migration estimates. The electoral data usually showed a much lower level of residential moves and very large fluctuations reflecting the timing of elections. The quality of electoral transfer data deteriorated after 1976 and by 1981 was no longer considered a reliable source of interstate migration statistics.

9.92 Because the combined scope of electoral data and family allowance data did not provide full coverage of the entire population, adjustments were made to allow for the persons not covered, namely persons over 15 years of age but below voting age, and others not eligible to vote.

9.93 The deterioration of the electoral change of address data led to their abandonment in 1981. From 1981 to 1986, family allowance data only were used but in October 1987, when the means test for family allowance payments was introduced, it was also replaced.

9.94 A detailed assessment of the quality of the family allowance data, the internal migration survey (a former annual survey on change of usual residence) and, the electoral roll transfer data, are presented in the Occasional Paper: Post-censal interstate migration estimates 1966-1981 (ABS & Di Iulio 1984).

9.95 As from 1986, change of address data became available from the nationwide compulsory Medicare system and they were assessed to be of high quality. They have been used since then, in place of family allowance data, to estimate interstate migration. Chapter 7 - Estimating interstate migration describes the method being used to generate interstate migration statistics from Medicare change of address data.

9.96 Until June 1996, only Medicare change of address data for those aged 1-14 were used, as most other ages had significant under-registration of interstate movements. These movements were then expanded with adult to child expansion ratios based on information from the latest available Census being applied to Medicare change of address data for ages 1-14 for each interstate flow.

9.97 From July 1996, estimates for interstate migration have used Medicare change of address data for all ages as a result of improvement in registrations. Medicare data for each age have been used to represent total movements for that age, after making adjustments for under-registration (i.e. applying expansion factors) by comparing Medicare data with that from the latest available Census. For both intercensal periods 1996 to 2001 and 2001 to 2006 expansion factors were applied to males age 16-29 and females aged 18-24 universally for each state and territory for both arrivals and departures.

9.98 For the 2001-06 period, Medicare data were also supplemented by Defence Force data to compensate for interstate movements of defence force personnel not in the Medicare change of address dataset.

Previous Page