3101.0 - Australian Demographic Statistics, Dec 2006
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 05/06/2007
Page tools: Print All

TECHNICAL NOTE MEASURING NET OVERSEAS MIGRATION, METHOD USED SEPTEMBER QUARTER 2001 TO JUNE QUARTER 2006

BACKGROUND

The Improved method for calculating NOM

1 The Australian Bureau of Statistics (ABS) has developed an improved method for calculating net overseas migration (NOM) for September quarter 2006 onwards. Estimates from the past time series and the current time series based on the improved method are not comparable. Preliminary estimates for September and December quarters 2006 based on the new method are included in Table 2 on page 21 of this issue. The key change is the introduction of a '12/16 month rule' for measuring a person's residency in Australia, replacing the previous '12/12 month rule'. For further information on the new method see Information Paper: Improved Methods for Estimating Net Overseas Migration (cat. no. 3107.0.55.003) and Information Paper: Statistical Implications of Improved Methods for Estimating Net Overseas Migration, Australia 2007 (cat. no. 3107.0.55.005).

The previous method for calculating NOM

2 The remainder of this technical note summarises the previous method for calculating NOM used between the September quarter 2001 and June quarter 2006. The time series using this previous method has finished at June quarter 2006.

3 Estimates of the Australian population are generated on a quarterly basis by adding natural increase (the excess of births over deaths) and net overseas migration (NOM) occurring during the period to the population at the beginning of each period. This is known as the cohort component method, and can be represented by the following equation:

P(t+1) = P(t) + B - D + NOM, where:
P(t) = the estimated resident population at time point t
P(t+1) = the estimated resident population at time point t+1
B = the number of births occurring between t and t+1
D = the number of deaths occurring between t and t+1
NOM = net overseas migration occurring between t and t+1.

4 For state and territory population estimates, an additional term is added to the equation representing net interstate migration occurring between t and t+1.

5 NOM accounts for around half of population growth at the national level. This note outlines how the ABS calculates NOM estimates by state and territory, including adjustments made to overcome some limitations of existing migration data.

6 The ABS estimates the level of NOM occurring during each quarter using data on incoming (i.e. arriving) and outgoing (i.e. departing) passenger movements at Australian air and sea ports. These movements are classified into three main categories depending on the stated duration of stay in Australia or overseas:

• permanent movement;
• long-term (one year or more) movement; and
• short-term (less than one year) movement.

7 Conceptually, NOM is the difference between permanent and long-term arrivals, and permanent and long-term departures. However, at the time a person crosses the Australian border, it is not empirically known how long they will actually spend in Australia or overseas. For example, overseas visitors might change their travel plans and extend their stay in Australia (perhaps utilising on-shore visa grants), or depart earlier than they first intended. Similarly, Australian residents travelling overseas may change their plans while abroad (e.g. some might state that they are departing the country permanently, but return less than a year later, while others might stay overseas longer than they initially intended).

8 Some of these differences between stated travel intentions and actual travel behaviour may also reflect short interruptions to longer periods of stay or absence. For example, overseas students arriving in Australia might state that they intend to stay for three years, but return home for brief periods during this time. Similarly, Australians working or studying overseas might state that they intend to be away for more than a year but return for brief holidays.

9 The following diagram summarises the contributions of different types of overseas movements to NOM. Estimates of NOM are derived from information provided on incoming and outgoing passenger cards, as well as other data supplied by the Department of Immigration and Citizenship (DIAC). Data on the intended duration of stay of overseas visitors arriving in Australia and the intended duration of absence of Australian residents travelling overseas are used to determine the numbers of permanent and long-term arrivals, and permanent and long-term departures. Passenger card data are also used to calculate migration adjustments and determine the state and territory distribution of NOM.

Migration adjustments

10 The ABS applies a number of adjustments to overseas arrivals and departures data in order to produce estimates of NOM. These mainly comprise adjustments designed to reflect differences between stated travel intentions and actual travel behaviour, but (in the case of revised NOM estimates) also include adjustments to transform numbers of overseas movements into numbers of travellers. These adjustments are collectively referred to as 'migration adjustments', although they have also been referred to in the past as 'category jumping' adjustments.

11 The processes of adjusting movement data on travellers' stated intentions to reflect their actual behaviour are complex, and depend upon the amount and type of movement data available at a particular point in time. The methods currently used compare data on actual travel movements over a one year period with those first advised by individual travellers, and are explained in more detail in Demography Working Paper 2003/5 - Net Overseas Migration: Adjusting for Actual Duration of Stay or Absence (cat. no. 3137.0). In order to conduct such a comparison, data for a 15 month period (i.e. one year plus one quarter) are required. These adjustment methods described in the working paper have been applied to NOM data for the September quarter 2001 onwards and will be subject to further investigation and improvement with the accumulation of additional data and time series.

12 Table 1 describes the impact that various types of migration adjustments have on NOM estimates. The adjustments applied to preliminary and revised NOM estimates are described in more detail elsewhere in this document.

 1. MIGRATION ADJUSTMENTS APPLIED TO NOM ESTIMATES Migration Adjustment Treatment in adjusted estimates ADJUSTMENTS MADE TO PRELIMINARY NOM ESTIMATES Persons whose stated travel intentions differed from assumed travel behaviour(a) Long-term visitor arrivals assumed to be staying in Australia short-term Subtract from NOM Long-term resident departures assumed to be staying overseas short-term Add to NOM Short-term visitor arrivals assumed to be staying in Australia long-term Add to NOM Short-term resident departures assumed to be staying overseas long-term Subtract from NOM ADJUSTMENTS MADE TO REVISED NOM ESTIMATES Persons whose stated travel intentions differed from actual travel behaviour(b) Permanent arrivals who actually stayed in Australia short-term Subtract from NOM Permanent departures who actually stayed overseas short-term Add to NOM Long-term visitor arrivals who actually stayed in Australia short-term Subtract from NOM Long-term resident departures who actually stayed overseas short-term Add to NOM Short-term visitor arrivals who actually stayed in Australia long-term Add to NOM Short-term resident departures who actually stayed overseas long-term Subtract from NOM Multiple movements of travellers Subtract from NOM(c) (a) Based on trends observed for the proportions of long-term and short-term arrivals and departures who change their travel behaviour. (b) Based on matched passenger records comparing stated travel intentions with actual behaviour. (c) Numbers of movements are converted into numbers of persons by matching passport numbers and other identifying personal details.

State and territory distribution of NOM

13 The state and territory distribution of NOM is based on information reported by travellers on arrival in or on departure from Australia. Incoming passenger cards provide information on the state or territory of a traveller's intended address within Australia, while outgoing passenger cards provide information on the state or territory in which a traveller lives or spent most time. However, the way in which this distribution is calculated differs between preliminary and revised estimates of NOM due to the amount of data available.

14 The following sections of this document describe how preliminary and revised estimates of NOM are created and distributed between states and territories. Estimates of NOM are finalised after the five-yearly Census of Population and Housing.

PRELIMINARY NOM ESTIMATES

15 The ABS produces quarterly estimates of Australia's resident population (known as the ERP) five to six months after the end of the reference quarter, and is required under legislation to provide population estimates as at 31 December by 6 June of the following year. Since estimates of NOM (adjusted for actual travel behaviour) require 15 months of data, preliminary estimates of NOM are calculated to meet more immediate ERP requirements.

Migration adjustments

16 There are four main groups of travellers who provide an intended duration of stay on their passenger cards who have the potential to change their duration of stay or absence:

• long-term overseas visitors who stayed in Australia for less than 12 months (i.e. long-term visitors who stayed in Australia short-term);
• short-term overseas visitors who stayed in Australia for 12 months or more (i.e. short-term visitors who stayed in Australia long-term);
• Australian residents departing long-term who stayed overseas for less than 12 months (long-term departures who stayed overseas short-term); and
• Australian residents departing short-term who stayed overseas for 12 months or more (short-term departures who stayed overseas long-term).

17 Migration adjustments applied to preliminary NOM estimates are based on the trends observed for the proportions of long-term and short-term arrivals and departures who change their travel behaviour. Table 2 shows the proportion of long-term and short-term travellers in 2004-05 who had changed their stated travel intentions. Preliminary migration adjustments are only applied to the four major movement categories (i.e. long-term visitor arrivals, short-term visitor arrivals, long-term resident departures and short-term resident departures).

 2. Changes in travel behaviour(a), Selected categories of movement(b) - September quarter 2004 to June quarter 2005 Long-Term Short-Term Arrivals Departures Arrivals Departures Period % % % % 2004 September 67.5 49.8 2.5 2.2 December 65.4 48.7 2.5 2.2 2005 March 69.9 53.8 3.4 2.9 June 66.4 51.0 2.6 2.2 Average 67.3 50.8 2.7 2.4 (a) Proportion of travellers whose actual duration of stay or absence differed from their stated intentions. (b) Based on stated intentions.

18 An average adjustment based on the most recent complete financial year for which 15 months of data exist is applied to each new quarter of movement data. For example, preliminary NOM estimates for the September quarter 2006 assumed that, based on the 2004-05 evidence, 67.3% of long-term visitor arrivals during the quarter would in fact stay in Australia for less than 12 months, while 50.8% of long-term resident departures would return to Australia within 12 months.

19 Table 3 shows how the preliminary NOM estimate for the June quarter 2006 was calculated.

 3. Components of net overseas migration, Original and adjusted estimates - June quarter 2006 Original Estimate Migration Adjustment(a) Adjusted estimate for preliminary NOM Initial category of movement no. no. % no. Permanent movement Permanent (settler) arrivals 32 440 . . . . 32 440 Permanent departures -16 010 . . . . -16 010 Long-term movement Visitor arrivals 38 728 -26 062 67.3 12 666 Resident arrivals 21 197 . . . . 21 197 Visitor departures -21 631 . . . . -21 631 Resident departures -23 733 12 064 50.8 -11 669 Short-term movement Visitors arrivals 1 165 708 31 929 2.7 31 929 Resident arrivals 1 093 606 . . . . . . Visitor departures 1 247 695 . . . . . . Resident departures 1 302 298 -31 098 2.4 -31 098 Net overseas migration 30 991 -13 167 . . 17 824 . . not applicable (a) Refer to table 1 in this Technical Note for further information on the migration adjustments applied to preliminary NOM estimates.

State and territory distribution

20 As noted in paragraph 13, the state and territory distribution of NOM is based on information reported by travellers on arrival in or on departure from Australia. However, at the time preliminary NOM estimates are calculated, information on the state or territory in which long-time arrivals will actually spend most time is not available because outgoing passenger cards for these persons have not yet been completed. State and territory distributions of long-term arrivals therefore refer to the state or territory of their intended addresses, as advised on incoming passenger cards. Similarly, state and territory distributions of permanent arrivals refer to their intended addresses as advised on incoming passenger cards, which may differ from the state or territory where they settle in the long-term.

21 The state and territory distribution of preliminary migration adjustments for a particular quarter is assumed to be the same as that of permanent and long-term arrivals in the same quarter. In practice, a national total is calculated for the migration adjustment. This is then distributed across the states and territories, by age and sex, using the distribution of permanent and long-term arrivals by state or territory of intended address. For example, since 24.0% of all permanent and long-term arrivals in the June quarter 2006 intended to live in Victoria, 24.0% of the total migration adjustment (-3,165) is also applied to this state. Table 4 shows components of net overseas migration for June quarter 2006 by state and territory.

 4. Components of net overseas migration, States and territories - June quarter 2006 NSW Vic. Qld SA WA Tas. NT ACT Aust.(a) Category of movement no. no. no. no. no. no. no. no. no. Permanent and long-term arrivals 34 598 22 199 16 296 5 153 11 060 705 786 1 562 92 365 Permanent and long-term departures 24 689 14 066 10 855 2 838 6 256 618 424 1 627 61 374 Migration adjustment -4 932 -3 165 -2 323 -735 -1 577 -100 -112 -223 -13 167 Net overseas migration 4 977 4 968 3 118 1 580 3 227 -13 250 -288 17 824 (a) Includes Other Territories - see paragraph 2 of the Explanatory Notes.

22 The current method of distributing the preliminary migration adjustment across states and territories is the same as that which has been previously used for preliminary category jumping estimates (see paragraph A3.24 of Demographic Estimates and Projections: Concepts, Sources and Methods (cat. no. 3228.0).

23 However, the ABS plans to review this method, with the prospect of applying a distribution method which allows for positive as well as negative adjustments for individual states and territories. In the interim, the preliminary estimates of NOM are subject to revision when more complete data are available.

REVISED NOM ESTIMATES

24 Preliminary estimates of NOM for a financial year are usually revised in the following March issue of Australian Demographic Statistics (cat. no. 3101.0). These revised NOM estimates use matched passenger records to calculate the actual duration of stay relating to overseas movements. Migration adjustments applied to revised NOM estimates are based on these matched data and include, in addition to the four major movement categories previously identified, a subset of movements relating to permanent arrivals and permanent departures:

• permanent (settler) arrivals who arrived in and left Australia in the same quarter, and did not return at any point during the 12 months following this arrival; and
• permanent departures who left and returned to Australia in the same quarter, and did not depart at any point during the 12 months following this departure.

25 Migration adjustments applied to revised NOM estimates also adjust for multiple movements of travellers (i.e. converting numbers of movements into numbers of persons).

26 The current methodology for these revised migration adjustments has been applied from the September quarter 2004 to June quarter 2005. Table 5 shows how revised NOM estimates were calculated for 2004-05.

 5. Components of net overseas migration, Original and adjusted estimates - 2004-05 Original estimate Migration adjustment(a) Adjusted estimate for revised NOM Initial category of movement no. no. no. Permanent movement Permanent (settler) arrivals 123 424 -7 334 116 090 Permanent departures -62 605 3 420 -59 185 Long-term movement Visitor arrivals 202 195 -137 287 64 908 Resident arrivals 101 301 . . 101 301 Visitors departures -94 707 . . -94 707 Residents departures -91 635 46 850 -44 785 Short-term movement Visitors arrivals 5 408 339 148 771 148 771 Residents arrivals 4 541 569 . . . . Visitors departures 5 457 870 . . . . Residents departures 4 591 198 -108 630 -108 630 Net overseas migration 177 972 -54 210 123 763 . . not applicable (a) Refer to table 1 in this Technical Note for further information on the migration adjustments applied to revised NOM estimates.

State and territory distribution

27 As is the case for preliminary NOM estimates, the state and territory distribution of revised NOM estimates is determined based on information reported on incoming and outgoing passenger cards (i.e. state or territory of intended address for arrivals and state or territory of residence/spent most time for departures).

28 The state and territory distributions of the migration adjustment are calculated based on the initial passenger card that identifies the movement of the traveller. For example, a long-term resident departure who returned to Australia within twelve months is added back to the state of residence they reported on departure (as identified on their outgoing passenger card). A long-term visitor arrival who actually stayed in Australia for less than twelve months is taken away from the state or territory they intended to live in (as identified on their incoming passenger card).

29 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 persons who originally stated that they intended to stay for less than twelve months, but actually stayed longer, and this component of the migration adjustment is treated differently.

30 In the absence of direct information from outgoing passenger cards for this group, the ABS has applied the state and territory distribution for short-term visitors departing Australia who were in Australia for between six and twelve months. The state and territory distributions used for revised NOM estimates (shown in table 6) are still subject to revision. The ABS expects that these estimates will improve as investigations proceed, and as actual data on state or territory of stay becomes available for this segment of the overseas visitor population (i.e. as outgoing passenger cards become available).

 6. Components of net overseas migration, States and territories - 2004-05 Category of movement NSW Vic. Qld SA WA Tas. NT ACT Aust.(a) Permanent and long-term arrivals 165 216 106 093 71 501 20 876 48 592 3 707 2 661 8 265 426 920 Permanent and long-term departures 102 567 56 546 43 131 10 757 24 440 2 457 1 847 7 190 248 947 Migration adjustment -27 444 -17 255 1 185 -3 099 -6 992 -205 190 -589 -54 210 Net overseas migration 35 205 32 292 29 555 7 020 17 160 1 045 1 004 486 123 763 (a) Includes Other Territories - see paragraph 2 of the Explanatory Notes.

CHANGES TO MIGRATION ADJUSTMENT METHODS

31 Due to changes in the methods used to adjust NOM estimates, caution should be used when comparing estimates over time. Table 7 describes the adjustment methods that have been applied to NOM estimates since September quarter 1996 (i.e. since the last intercensal period). Adjustments applied to overseas migration estimates have also been discussed in a special article in Migration, Australia, 2002-03 (cat. no. 3412.0).

 7. Migration adjustment methods - September quarter 1996 to June quarter 2006 Period Adjustment method September 1996 - June 1997 Category jumping' adjustments applied using previous methodology(a) September 1997 - June 2001 No adjustments applied (i.e. 'category jumping' set to zero) September 2001 - June 2005 Current migration adjustments used (revised NOM estimates) September 2005 - June 2006 Current migration adjustments methods used (preliminary NOM estimates) (a) For further information, refer to Appendix 3 in Demographic Estimates and Projections: Concepts, Sources and Methods (cat. no. 3228.0).

FURTHER INFORMATION

32 For further information on the measurement of NOM, contact Phil Browning on Canberra (02) 6252 6639.