3101.0 - Australian Demographic Statistics, Jun 2003
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 11/12/2003
Page tools: Print All

Feature Article - Overseas Migration: People whose stated length of stay is different from their actual length of stay

ABSTRACT

In order to calculate the change in Australia's official population measure, Estimated Resident Population (ERP), apart from natural increase (the excess of births over deaths) the other key component is Net Overseas Migration (NOM).

The ABS has revised Net Overseas Migration data for 2001 - 02 financial year and released a preliminary modelled estimate for 2002-03 financial year based on actual traveller information. The level of adjustment to the Australian NOM in 2001-02 is final and not subject to review, though the State distribution may be reviewed as more information becomes available over time. Preliminary adjustments to NOM are subject to revision as more data becomes available.

This article examines the concepts of NOM, the results and methodology for estimating NOM.

NET OVERSEAS MIGRATION CONCEPTUAL FRAMEWORK

The concept of NOM is simply permanent and long-term overseas arrivals minus equivalent departures. A person who spends 12 months in Australia is added to the population while a person who spends less than 12 months in Australia is not added to the population for that reference period. A resident of Australia who departs for 12 months is taken out of the Australian population whereas if they depart for less than 12 months, they are not taken out. This is also known as the '12 month residency rule', or the criterion for the length of time needed for being counted in Australia's ERP. To date, the 12 month residency rule has been applied as an unbroken period.

In its simplest form, NOM can be shown as:

(1) NOM = A - D

where A is the arrival of a person not already included in the Australian population and D is the departure of a person who is already included in the population in the reference period under consideration. For the purposes of calculating NOM, arrivals and departures are broken down from (1) above into:

A = permanent arrivals + long-term arrivals

D = permanent departures + long-term departures

Conceptually then, net overseas migration is simply the difference between permanent and long-term arrivals and permanent and long-term departures.

(2) NOM = Permanent Arrivals - Permanent Departures
+ Long-term Arrivals - Long-Term Departures

Information about traveller movements is accumulated each calendar month. A revised NOM based on actual duration of stay in or absence out of Australia is calculated over the 15 months after the reference quarter. Due the timeliness required for NOM data, a means of estimating preliminary NOM is needed. Preliminary NOM is calculated from traveller movements data for the current quarter and adjusted by a model which converts movements into people based on historical differences between movements and actual duration of stay or absence in traveller categories.

LENGTH OF STAY

Initially length of stay is recorded from a travellers 'intended length of stay or absence' from the passenger cards when they first crossed the border. This forms the basis for NOM and is then adjusted for the number of travellers that change from their intended length of stay in Australia.

At the time a person crosses Australia's border, it is not empirically known how long they will actually spend in Australia (eg for an arrival). For a departure, it is not known how long they will actually spend overseas. Actual duration of stay or absence for the purposes of adding or subtracting from ERP is not know until up to 15 months after the reference quarter. For instance, a traveller entering Australia with the intention to stay for more than 12 months in the September quarter can not be confirmed as a resident for the purposes of ERP until the end of the September quarter in the following year.

Actual duration of stay is calculated by matching passenger movements over the 12 months after they moved across Australia's borders. If a movement is not registered over the 12 months after a traveller has arrived or a resident has departed then they are added (in the case of visitors) or subtracted (in the case of residents departing) from Australia's ERP.

ADJUSTING NET OVERSEAS MIGRATION FOR TRAVELLER CHANGE IN INTENDED BEHAVIOUR

As mentioned previously, NOM is adjusted for those people who cross Australia's border and whose actual duration of stated length of stay either in Australia or overseas is different from their intended duration.

There are four main categories of traveller that add or subtract to Australia's ERP. Table 1 shows the categories of traveller that impact on Australia's ERP and are not permanent movements to Australia. There is an element of permanent arrivals and departures whose actual movement differs from their intention to be a permanent migrant, however, these are relatively small in size compared to the other categories.

TABLE 1 HOW DIFFERENT TRAVELLER TYPES ARE TREATED
 Category based on stated intention Population adjustment Short-term visitor arrival who stayed long-term Add to population Long-term visitor arrival who stayed short-term Subtract from population Short-term Australian resident departure who stayed overseas long-term Subtract from population Long-term Australian resident departure who stayed overseas short-term Add to population

Since September quarter 1997 ABS had set any adjustments to NOM to zero because of concerns with data quality and the methodology needed to calculate revised NOM. See Demography Working Paper 2003/1 Estimated Resident Population and the Measurement of Category Jumping. In Australian Demographic Statistics, June 2003 quarter (cat no. 3101.0) a revised NOM was released for 2001-02 financial year and a preliminary NOM for the 2002-03 financial year.

Revised NOM estimates are presented in Table 2 to illustrate the adjustment made to NOM up to 15 months after the reference quarter for each quarter over the financial year 2001/2002. It can be noted that not every quarter has a negative revision. Due to the method used to calculate an adjustment to NOM, some quarters will have a positive revision due to the way that some categories are now behaving over the December quarter. More information can be found in Demography Working Paper 2003/5 Net Overseas Migration: Adjusting for Actual Duration of Stay or Absence.
TABLE 2 ESTIMATES OF REVISED NET OVERSEAS MIGRATION
 Migration components September qtr 2001 December qtr 2001 March qtr 2002 June qtr 2002 2001-2002 Permanent arrivals no. 22,833 21,348 22,163 22,556 88,900 Permanent departures no. 11,632 11,222 14,449 10,938 48,241 Net permanent migration no. 11,201 10,126 7,714 11,618 40,659 Long-term arrivals no. 65,926 59,021 91,783 47,741 264,471 Long-term departures no. 41,477 41,695 48,909 39,365 171,446 Net long-term movements no. 24,449 17,326 42,874 8,376 93,025 NOM adjustment no. -7,938 3,737 -14,233 -4,694 -23,128 Revised net overseas migration(a) no. 27,712 31,189 36,355 15,300 110,556 % difference of final NOM from published NOM % -22.3 13.6 -28.1 -22.5 -17.3 (a) includes (i) an element of adjustment for multiple times inclusion/exclusion in the same quarter, (ii) an element of adjustment of people arriving and departing permanently, and (iii) the four groups of people providing intended duration of stay on the passenger card.

NEW METHODS

Adjusting for revised Net Overseas Migration

From September quarter 2001 to June quarter 2002, ABS has implemented a new method of measuring an adjustment to NOM based on matched movement histories for people who crossed Australia's border. The matching of movements occurs over the 24 month period between July 2001 and July 2003. This allows the actual duration of stay to be calculated for travellers and for a 12 month residency criteria to be applied to their movements (see Net Overseas Migration conceptual framework section above).

Net Overseas Migration for the year 2001-02 was calculated from intended movements as recorded from passenger card information adjusted by those whose actual travel behaviour was different to their notified intended behaviour after their first movement in a quarter. This methodology takes into account of those travellers who have particular requirements to move multiple times within a quarter such as long-term business migrants who are using Australia as a regional base. These movements are taken out of NOM for that quarter of initial matched movement.

Adjusting for preliminary Net Overseas Migration

For the 2002-03 financial year and onwards, a preliminary estimate of those travellers who may change their intended behaviours is applied to NOM. This estimate will be revised once data are available up to 15 months after the reference quarter.

The estimation method for preliminary Net Overseas Migration is still under review as more information becomes available. The current method is to average the four previous known quarters of adjustments and apply these to the next four quarters. At this time this is the most stable means for calculating a preliminary estimate of NOM.

Over time more factors will be examined to attempt to enhance the preliminary estimate of NOM. The two most volatile components of NOM are short-term visitors who stay for more than 12 months and those Australian residents who leave for a short-term overseas trip and stay for more than 12 months. While the current estimation method for these types of traveller is the best available, more work is needed in the future as more data becomes available. More information about the estimation method can be found in Demography Working Paper 2003/5.

State distribution

The state or territory distribution of permanent or long term migration is determined based on information reported on arrival or departure from Australia.

This method may be considered to be reasonable for people who, on arrival, intend to settle or stay in Australia for more than twelve months. However, there is less certainty about the reliability of the state or territory of intended stay for those who originally stated that they were intending to stay for less than twelve months, but in fact stayed longer.

For preliminary estimates of NOM, the state/territory distribution has been derived using the distribution of permanent and long term arrivals.

In the absence of direct information from outgoing passenger cards for this group, the ABS has used information of actual state or territory of stay distribution from short term visitors departing Australia who were in Australia between six and twelve months. The state or territory distributions of NOM are still subject to revision and it is expected that these estimates will improve as investigations are undertaken over financial years as actual data on state or territory of stay becomes available for this segment of the overseas visitor population when they leave Australia.

More information can be found in the following documents or by contacting Phil Browning on (02) 6252 7612 (phil.browning@abs.gov.au) or Abul Majumder on (02) 6252 7083 or a.majumder@abs.gov.au:

BIBLIOGRAPHY

Demography Working Paper 2003/1 Estimated Resident Population and the Measurement of Category Jumping
Demography Working Paper 2003/5, Net Overseas Migration: Adjusting for Actual Duration of Stay or Absence