Australia's demographic future: determinants of our population

Transcription

1 Australa's demographc future: determnants of our populaton Chrstabel Young and Lncoln Day Durng the past fve years or so there has been a change n atttude about Australa's populaton, from concern about whether t was growng too slowly, to a questonng about whether t s growng too fast. Smlarly, there has been a change from concern that Australa's brth rate was too low, to some relef that t s near replacement level. n addton, whle hgh mmgraton durng the post-war years was actvely promoted as a means of achevng rapd populaton growth, there s now recognton that even a low level of annual net mgraton could result n a future populaton n excess of 25 mllon (even wth fertlty contnung at 10% below replacement level). What can demographers tell about the future populaton? Demographers cannot decde what the optmum populaton sze should be that s the task of the envronmental scentsts but they can show the effect on future populaton sze of a gven combnaton of fertlty, mortalty and net mgraton. Accordngly, a major am of ths paper s to dscuss feasble or attanable levels of future populaton sze by descrbng the effect of feasble ranges of fertlty, mortalty and net mgraton. 19

2 Populaton 2040 Demographers can also work out what level of annual net mgraton s consstent wth a gven populaton target. For example, some poltcans have suggested 25 mllon persons as the deal populaton for Australa. Senator Robert Ray, who was Mnster for mmgraton at the tme, stated ths fgure on Channel 10's 'Face to Face' n md-august 1989, and Bob Hawke, who was Prme Mnster at the tme, gave the same fgure at the Natonal mmgraton Outlook Conference n November Demographers were able to show that, wth a contnuaton of the current level of 10% below-replacement fertlty, ths target could be acheved and mantaned wth a constant net mgraton ntake of only 70,000 persons per year. Although demographers cannot say how bg an optmum populaton should be, they nonetheless can specfy the desrable demographc characterstcs of an optmum populaton. n partcular, the deal populaton would have: (a) low mortalty; (b) a farly regular age structure, so as to avod the socal and economc upheaval resultng from bg dfferences between numbers n successve age cohorts; and (c) a near-zero growth rate. deally, also, the optmum populaton sze, whatever t s, should be attaned gradually nstead of watng untl crss condtons forced the ntroducton of drastc measures to reduce populaton growth (also see Day, 1971). Methodology, data and defntons To answer these questons about the future populaton, use s made of sets of populaton projectons, derved by the Australan Bureau of Statstcs (ABS) (1990, 1994b), Young (1990a, b) and Day (1994). These have all been carred out usng the standard cohort-component method of projecton, n whch age-specfc fertlty rates, sex/age-specfc mortalty rates, and a sex/age structure of the mgrant ntake (usually the age structure of the net ntake over the past few years) are appled to the sex/age structure of the populaton at a gven startng date, such as a census. The projectons by ABS and Young use varous levels of constant annual net mgraton, whle the projectons by Day use two sets of age-specfc mgraton rates. Some of the terms used by demographers may requre some explanaton: Natural ncrease = annual number of brths mnus annual number of deaths. Rate of natural ncrease = the crude brth rate mnus the crude death rate (the crude rate beng the rato of the annual number of brths or deaths to the total populaton, usually expressed per 1000 of the populaton). The net reproducton rate (NRR) s the average number of daughters that a woman would bear throughout her reproductve years accordng to exstng condtons of fertlty and mortalty. 20

3 Chnstabel Young and Lncoln Day The total fertlty rate (TFR) s the average number of chldren that a woman would bear durng her reproductve lfetme accordng to exstng fertlty rates. Replacement level fertlty s represented by a NRR of 1.00, and n a low-mortalty country such as Australa, a TFR of However, a fertlty level below replacement wll only result n a declne n populaton sze f the populaton has a regular age structure, as a result of a long hstory of constant fertlty and mortalty (.e. a statonary populaton). n Australa, where there s a large cohort at the adult ages, natural ncrease contnues to be postve (.e. the annual number of brths exceeds the annual number of deaths) even though fertlty has been belowreplacement level snce A statonary populaton has a constant sze and a fxed age/sex structure. t s the result of a long perod of constant numbers of brths (from replacement level fertlty) and constant age-specfc mortalty rates. Alternatvely a statonary populaton can also result from a constant level of below-replacement fertlty, constant age-specfc mortalty rates, and a constant level of annual net mgraton (equal to the defct of brths n relaton to deaths). The nfluence of age structure The domnant feature of Australa's current populaton age structure, compared wth a statonary populaton relatng to the current level of mortalty, s the defct n the number of older persons aged 50 years and over, and the excess n the number of persons aged 0-44 years (see Fgures 1 and 2). Ths s due to the combnaton of the post-war baby-boom and the contnung hgh numbers of brths snce then. Whle most European countres had a dstnct post-war baby boom for a few years and then a return to lower numbers of brths, n Australa, numbers of brths snce the 1950s have remaned near the level durng the boom perod. Ths has resulted n an unusually large proporton of the populaton at both the chldbearng ages (and hence an excess of brths over deaths despte belowreplacement fertlty) and the workng ages (and hence an oversupply of workers), a condton whch wll contnue for another 15 years or so. Subsequently, from 2011 to 2031, as these baby boomers reach ther sxtes, there wll be a rapd ncrease n the proporton of elderly persons n the populaton. 21

4 Populaton LZ 1 " 1 ZD 1 E_ ZD zn : Z3 1 1 l 1 j,,, 1, 3.00 MALES Populaton (%) 3.00 FEMALES Source: Based on data from Australan Bureau of Statstcs (1994a) Fgure 1: Australa. Populaton 30 June 1992 S* 104) S-S cz r_ MALES Populaton (%) FEMALES Source: Based on data from Australan Bureau of Statstcs (1993) Fgure 2: Australa. Statonary populaton accordng to the 1992 lfe table (constant brths, constant deaths, no mgraton) 22

5 Chrstabel Young and Lncoln Day Feasble lmts of mortalty, fertlty and net mgraton Mortalty Age-specfc mortalty rates measure the underlyng experence of mortalty, thus determnng the proporton of the populaton survvng through chldhood, adulthood and nto old age. n Australa these rates have generally been declnng over tme. Although the usual assumpton s that age-specfc mortalty levels n Australa wll contnue to declne, some commentators beleve that they could ncrease n response, for example, to worsenng envronmental condtons n some of Australa's major urban areas, marked dfferentals accordng to socoeconomc status, or ncreased acceptance of euthanasa. Certanly, wth Australa's agespecfc mortalty rates already low, any future reductons n mortalty are lkely to be small. Government nterventon wth regard to mortalty has always been to reduce age-specfc mortalty rates and to ncrease well-beng. The only foreseeable departure from ths n the future would be a relaxaton of the present strctures on euthanasa (dscussed n Day, 1992). Day (1994) shows that a 10% ncrease n mortalty between 1992 and 2037 at ages 70 years and over would make a dfference of less than 400,000 n the sze of the total populaton n 2042 (18.8 mllon compared wth 19.2 mllon followng zero net mgraton snce 1992). There would also be somewhat lower proportons of elderly n such a populaton as a result of ths ncrease n mortalty at the old ages, 20.0% compared wth 21.4% at ages 65 years and over; 8.7 compared wth 10.3% at ages 75 years and over; and 4.7 compared wth 5.9% at ages 85 years and over n The crude death rate s the rato of total deaths to the total populaton, and s relevant to overall populaton change. The crude death rate generally ncreases as a populaton ages, even f age-specfc mortalty rates are constant or declnng. As ts populaton gradually becomes older n the comng decades, the crude death rate n Australa s expected to ncrease gradually. Ths, n turn, wll gradually reduce the rate of natural ncrease. Government can have lttle effect on the crude death rate, as ts expected ncrease s a functon of the nevtable ageng of the populaton. A declne n mortalty results n an ncrease n the proporton of the populaton survvng to ages 65 years and beyond. Some studes suggest that these addtonal years of lfe ganed n old age are mostly years of llness and dsablty, leadng to concern about the hgher costs of health care of the elderly populaton. Ths, together wth the predcted level of government expendture on age pensons for an ncreasng number of elderly persons, has produced the vew that there s an ageng problem. However, to the extent that there s such a problem, t s a functon not only of the proporton of the elderly, but of such other factors as ther socoeconomc characterstcs, ther health status, level of educaton, employment 23

6 Populaton 2040 hstory, superannuaton cover, home ownershp, and the presence of grownup chldren to provde care and other support. Fertlty When commentators talk about the trend n fertlty they are usually referrng to annual measures, such as the net reproducton rate (representng the number of survvng daughters per woman accordng to a gven year's fertlty experence), or the total fertlty rate (representng the average number of chldren per woman accordng to fertlty experence durng a gven year). t s very dffcult to predct the future level of fertlty, but feasble lmts for the long-term trend would appear to be between 10 to 20% below replacement and 10% above. Although Australa's level of fertlty has been relatvely constant at around 10% below replacement over the past 15 years snce 1978 (see Fgure 3), before then t was very erratc, and t could well become erratc agan n the future. 1.8T oc % c o o 3 o o a. 0) 0) Z 0.8 H Year Source: Based on data from Australan Bureau of Statstcs, Cat. No Fgure 3: Annual net reproducton rate (NRR), 1950 to 1992 Annual fertlty rates can vary wdely as a result of changes n the tmng of brths by couples, even f each successve generaton of women has the same average number of chldren. For example, the post-war baby boom was largely due to earler marrage and earler tmng of the frst brth, whle the declne n the 24

7 Chrstabel Young and Lncoln Day fertlty rate durng the 1970s was largely due to delayed marrage and later chldbearng. f, for example, Australan women took the recent medcal advce that they should begn chldbearng at age 19 years to avod breast cancer, we would see a rapd rse n annual fertlty rates and another baby boom, subsequently followed by a compensatng declne n annual fertlty, even f the women themselves had the same ultmate number of chldren. The debate n early 1994 about the legalty of aborton n New South Wales could also have a temporary mpact on the annual number of brths. Because annual fertlty n Australa s already below replacement level, t would be napproprate to try to force t lower. Moreover, as suggested n the dscusson about an optmum populaton, forcng fertlty too far below replacement could produce severe dstortons of the age structure of the populaton. t would also be napproprate to try to reduce fertlty further below replacement level whle there s a large ntake of mmgrants, especally when so many of them are from hgh-fertlty countres and are able to brng n large numbers of chldren, and to sponsor sblngs from large famles under the famly reunon program. f fertlty ncreased to above replacement level, one opton for haltng excessve growth could be to reduce government assstance, such as famly allowance payments, wth respect to, say, fourth or hgher-order brths. Obvously, there would need to be forewarnng of such legslaton, so as not to penalse chldren already born or conceved, and also wth a specal dspensaton for multple order brths. Under the current stuaton of below-replacement fertlty n Australa, fourth or hgher order brths account for only about 7.5% of all brths, but the assumpton s that these mght represent a hgher proporton f overall fertlty were hgher. Reducng assstance to thrd-order brths would seem to be too drastc a measure. There s a vew that, because the average number of chldren s hgher among low than among mddle or upper ncome famles (Hugo, 1993), famly allowance payments, whch are strongly focused on low ncome famles, could be havng a mldly pronatalst effect. f ths s so, then a reducton n ncome support to fourth or hgher-order brths could have some mpact. mmgraton Throughout ts hstory, Australa has experenced large fluctuatons n ts mmgraton ntake. Durng the last 15 years, snce 1978, annual net mgraton has vared from 47,000 to 172,000, and the annual net mgraton rate has vared from 0.33 to 1.05%. The man effect of mmgraton s to add to the sze of the populaton. A year's ntake of mmgrants adds to the populaton n two ways: the mmgrants 25

8 Populaton 2040 themselves and ther chldren who are subsequently born n Australa (or ther share, n the case of ntermarrage); and there s the subsequent demand to brng relatves to Australa as part of the Famly Reunon program. A constant ntake of 50,000 net mgraton for 50 years to 2040 wth 10% below-replacement fertlty would generate a populaton of 3.3 mllon, one-thrd of whch would be a result of ther chldren born snce arrval. f the level of fertlty were hgher or f the age structure of each year's mgrant ntake were younger than s currently the case, then the ncrease generated would be even larger. Studes n Australa (Young, 1989, 1990a), Canada (Denton and Spencer, 1987; Beaujot, 1989), and the Unted States (Coale, 1987) confrm that mmgraton wll do very lttle to make the populaton less aged. Although a year's ntake of mmgrants mght add to the labour force, a few decades later those mmgrants age and enlarge the numbers of elderly. The small effect of net mgraton n retardng the ageng of the populaton s also at the expense of a much larger ncrease n the total populaton, as Table 1 ndcates. Table 1: Effect of fertlty and net mgraton on the proporton of persons aged 65 years and over n relaton to changes n the sze of the total populaton 50 years ahead Projectons from 1992 to 2042 Rate of annual net mgraton* Dfference Startng at 0.33% Startng at 0.85% Fertlty Remans at 10% below replacement % 65 years+ Total populaton (mllons) ncreases to replacement level n 2002 % 65 years+ Total populaton (mllons) Dfference % 65 years Total populaton (mllons) * Day's (1994) projectons use age-specfc rates of net mgraton, and so the overall rate may change durng the course of the projectons. 26

9 Chrstabel Young and Lncoln Day When the level of fertlty remans at 10% below replacement, Day's (1994) projectons show that a hgh rate of annual net mgraton (0.85%) would result n 17.4% of persons beng aged 65 years and over n 2042 compared wth 19.9% wth a moderate rate of annual net mgraton (0.33%). However, at the same tme, the populaton attaned n 2042 wth the hgh rate of mmgraton would be 8.8 mllon greater than that acheved through moderate mmgraton (32.5 mllon compared wth 23.7 mllon) (see Table 1). Smlar results occur f fertlty ncreases to replacement level n 2002; n ths case the hgher level of mmgraton results n 15.9% of persons aged 65 years and over compared wth 18.2% followng moderate mmgraton (a dfference of 2.3 percentage ponts). But agan ths s acheved at the expense of an addtonal 9.7 mllon persons n 2042 (35.5 mllon compared wth 25.8 mllon) (see Table 1). Earler fgures by Young (1990a) confrm that the small mpact of hgh levels of mmgraton n retardng of the ageng of the populaton s overshadowed by the very large ncrease n the total populaton whch accompanes ths. For example, ncreasng fertlty from 10% below replacement to replacement level (or by an average of one-ffth of a chld per famly) has the same effect n retardng the ageng of the populaton over a 60 year projecton perod as ncreasng annual net mgraton by 100,000, but wth less than one-thrd of the ncrease n populaton. Therefore, f the am were to retard the ageng of the populaton, a small ncrease n fertlty would be more effcent than the necessarly large ncrease n mmgraton. The mpact of feasble changes n, and current levels of, each component on the future populaton The varous populaton projectons prepared by the Australan Bureau of Statstcs, Young and Day llustrate the relatve effects of feasble ranges of the three components, mortalty, fertlty and net mgraton on the sze of the populaton around the year For example, n Table 2, Day's (1994) fgures show that wth moderate net mgraton (0.33% per year) the populaton n 2042 would be 2.1 mllon larger f fertlty rose to exact replacement level by the year 2002 than f t remaned at 10% below replacement level. Wth hgher net mgraton (0.85%), the same ncrease n fertlty would result n a dfference of 3.0 mllon. But the ncreases n total populaton produced by the hgher as aganst the moderate rate of net mgraton would be substantal: an ncrease of 8.8 mllon f fertlty remaned at 10% below replacement, and 9.7 mllon had fertlty ncreased to replacement by Note that an ncrease to replacement fertlty has a greater mpact on the populaton when net mgraton s hgh. Smlar results are also found wth regard to feasble changes n constant levels of annual net mgraton (n the second half of Table 2). 27

10 Populaton 2040 Table 2: Effect of the combnaton of feasble changes n levels of fertlty and net mgraton on the sze of the total populaton 50 to 60 years ahead Projectons from 1992 to 2042 (Day, 1994) Fertlty Rate of annual net mgraton* Startng at 0.33% Startng at 0.85% Mllons Dfference Remans at 10% below replacement ncreases to replacement level n Dfference Projectons from 1986 to 2046 (Young, 1990a) Fertlty Remans at 10% below replacement Constant annual net mgraton 50, ,000 Mllons Dfference 9.5 mmedately ncreases to replacement level Dfference * Day's (1994) projectons use age-specfc rates of net mgraton, and so the overall rate may change durng the course of the projectons. The level of fertlty has remaned at 10% below replacement snce The rate of annual net mgraton has vared from 0.33% to 1.05% snce 1978, or from 47,000 to 172,000. Projectons by the Australan Bureau of Statstcs (1990) from 1989 to 2031 wth TFR = 1.78 and annual net mgraton of 125,000, show that the 2031 populaton would be 26.0 mllon wth an expected declne n age-specfc mortalty rates, and 24.9 mllon wth constant mortalty rates a dfference of 1.1 mllon 28

11 Chrstabel Young and Lncoln Day Fgure 4 also shows the major effect of feasble rates of net mgraton compared wth feasble levels of fertlty. n contrast, the range n the future populaton n 2042 f mortalty decreased at all ages nstead of remanng constant s lkely to be just over one mllon (ABS, 1990). The potental range of the future level of mmgraton obvously has the largest mpact on the sze of the future populaton. G. FHHH L M 0 M2M 0 HMO L M M M2M M H M M L M H M2W H H M H Net mgraton equals zero Total populaton (mllons) Rate of annual net mgraton around 0.33% FERTLTY L : 10% below replacement M2 : reaches replacement n 2002 H : reaches replacement n 2002, and 10% above n 2037 Rate of annual net mgraton around 0.85% Projectons by Day (1994) usng age-specfc rates of net mgraton Fgure 4: Projected populaton n 2042 accordng to varous scenaros of fertlty and net mgraton A common approach s to assess the effect of a constant level of net mgraton, and ths s done n the offcal populaton projectons prepared by the Australan Bureau of Statstcs (ABS) (1994b). The most recent ABS projectons span the years 1993 to 2041, assume a gradual declne n mortalty, and are avalable for three levels of net mgraton (zero, an ncrease to 70,000 n , and an ncrease to 100,000 per year n ), and for three levels of fertlty (a contnuaton of the current 10% below-replacement fertlty, an ncrease to replacement level n 2002, or a decrease to 20% below replacement n 2002). These projectons show that wth 10% below-replacement fertlty the populaton n 2041 would be 20.3 mllon wth zero net mgraton, 24.9 mllon f annual net mgraton ncreased to 70,000, and 26.7 mllon f annual net mgraton ncreased to 100,000. The combnaton of replacement level fertlty and zero net mgraton would produce a populaton sze of 21.9 mllon n 2041, whle replacement level fertlty combned wth 70,000 net mgraton per year would produce a populaton of 26.7 mllon at the same date (see Table 3). 29

12 Populaton 2040 Table 3: Projected populaton at 2041 accordng to varous levels of constant annual net mgraton Constant level of annual net mgraton Populaton n 2041 Trend durng 2041 to 2051 Projectons by the Australan Bureau of Statstcs (1994b) Net Reproducton Rate , , Net Reproducton Rate , , Projectons by Young (1990b) Net Reproducton Rate 0.9; net mgraton below feasble lmts , , , , Net Reproducton Rate 0.9; net mgraton wthn feasble lmts 50, , , , , , , , , , Decrease Decrease Decrease Relatvely steady Relatvely steady Relatvely steady ncrease ncrease ncrease ncrease ncrease ncrease ncrease ncrease ncrease Projectons by the Australan Bureau of Statstcs (1994b) cover the perod 1993 to Projectons by Young (1990b) are a revson, made n 1990, of projectons publshed n Young (1990a). Assumptons: a gradual declne n age-specfc mortalty rates. Populaton at 30 June 1993 = 17.5 mllon. The range n net mgraton snce 1978 has been annual levels of 47,000 to 172,000, and annual rates of 0.33% to 1.05%. 30

13 Chrstabel Young and Lncoln Day An earler set of projectons by Young (1990b), based on less up to date nformaton, produce total populatons n 2041 whch are about one mllon less than the ABS (1994b) fgures. However, these projectons are useful to explore the effect of several dfferent levels of constant annual net mgraton on the future populaton (see Table 3 and Fgure 5). From Young's projectons, wth a contnuaton of the current 10% below-replacement fertlty, and a gradual declne n age-specfc mortalty rates, levels of annual net mgraton of 0 to 20,000 would result n frst an ncrease and then a slght decrease n total populaton before 2041, wth a gradual decrease thereafter. Wth constant annual net mgraton of 30,000 to 50,000, the populaton would ncrease up to 2041 and reman relatvely steady durng the next decade. Levels of 60,000 or more net mgraton per year would result n a contnuous ncrease to 2041 and nto the next decade. Only the constant level of 50,000 net mgraton s both wthn the feasble range (.e., wthn the range of actual net mgraton durng the past 15 years) and small enough to result n no ncrease n the populaton n the decade after 2041 (wth a contnuaton of the current level of 10% below-replacement fertlty). ~ , , , ,000 90,000 70,000 50,000 30,000 10,000 E O T- _ o o Source: Young (1990b) (Assumng a contnuaton of the current 10% below-replacement level of fertlty) Fgure 5: Projectons of the total populaton n Australa accordng to dfferent levels of annual net mgraton snce

14 Populaton 2040 Table 4 summarses the effect that feasble changes n the levels of mortalty, fertlty and annual net mgraton would have on the sze of the future populaton. t also ndcates the prospects for government nterventon to acheve any change. A feasble declne n age-specfc mortalty rates or a feasble hgher level of fertlty would each result n a slghtly larger populaton n 2040 than would a contnuaton of the current levels of mortalty and fertlty. But a feasble hgher level of net mgraton would result n a much larger populaton than that whch would occur wth a contnuaton of the current level of net mgraton. Overall the rato s 1:2.5:9.5. Table 4: Effect of feasble changes n mortalty, fertlty and net mgraton on the sze of the total populaton n 2041, and the possblty of government nterventon n reversng such changes Feasble changes (whch would produce hgher rates of populaton growth) Effect on the sze Relatve mpact Possblty of of the total (mllons) government populaton n 2041 compared wth a contnuaton of current levels nterventon Lower age-specfc mortalty rates Hgher age-specfc fertlty rates Hgher net mgraton Slghtly larger Slghtly larger Larger 1 None, because polcy s to reduce mortalty 2.5 Lttle 9.5 Major t s fortunate that the possblty for government nterventon s least for the two components (brths and deaths) whose feasble varaton has relatvely lttle effect on future populaton sze, and s greatest for the component (mmgraton) whose feasble varaton has the bggest effect on future populaton sze. Now let us look at the expected outcome n the future wthout any delberate attempt by government to nfluence the components of demographc change. Nether mortalty, whch s already very low, nor fertlty, whch s largely the result of decsons by couples, s lkely to change f the government wthdrew ts already slght nfluence over the levels of these two components. However, annual net mgraton s lkely to ncrease sgnfcantly wthout government nterventon. As Table 5 shows, over the next 50 years, natural ncrease (whch equals brths mnus deaths), the component over whch government has lttle control, wll gradually declne of ts own accord, even f fertlty ncreases to replacement level 32

15 Chrstabel Young and Lncoln Day (provded that net mgraton remans below a certan lmt). Ths s because the number of deaths s expected to ncrease as a result of the future ageng of the populaton. Therefore, although annual numbers of brths would reman relatvely constant (wth a contnuaton of near-replacement fertlty), natural ncrease would declne because of the ncrease n the annual number of deaths. n contrast, annual net mgraton s unlkely to reman stable, let alone declne of ts own accord and wll requre government nterventon to do so. Thus agan, t s fortunate that the two components over whch the government has lttle control, brths and deaths, combne to produce a declne n the annual ncrease n populaton. n contrast, the component whch has no nbult ablty to declne of ts own accord, s and can be controlled by government. Table 5: The effect of a contnuaton of the current stuaton, and the expected outcome n terms of populaton dynamcs Component of Contnuaton of current Expected outcome n terms of populaton change stuaton populaton dynamcs Mortalty Low age-specfc mortalty rates, but ncreasng ^ Combne to produce a annual deaths declnng natural ncrease Fertlty 10% below replacement J Natural ablty to make the fertlty, and near constant annual brths populaton grow more slowly Net mgraton Moderate now, but may No natural tendency to declne, ncrease but major government nterventon possble To summarse, mmgraton dffers from fertlty and mortalty, the other two components of change, n that: (a) the feasble range of annual net mgraton, taken as the maxmum and mnmum levels over the past decade or so, produces a wde range of populaton sze n the future, and far wder than that produced by a feasble range of fertlty or mortalty; and (b) t s unlkely to declne of ts own accord; but (c) t s responsve to government nterventon. So, (a) the natural ablty of Australa's populaton to grow more slowly; and (b) the ablty of the government to control the component most lkely to accelerate populaton growth would seem to suggest that there s no problem. But ths s not so, because there are powerful lobby groups demandng hgher levels of mmgraton, and because the responsblty for populaton matters n Australa les wth the Department of mmgraton, whose man functon and nterest s n mmgraton. 33

16 Populaton 2040 Australa's offcal polcy regardng populaton A look at Australa's offcal vew towards populaton hghlghts the problem. Ths offcal vew s conveyed n the recently publshed document, Australa. Natonal Report on Populaton (Natonal Commttee, 1994), wth the text of the report coordnated by the Department of mmgraton and Ethnc Affars. (Ths document wll be referred to as the Natonal Report n ths paper). The two paragraphs relatng to populaton polcy and optmum populaton are: The desrablty and character that a formal populaton polcy mght take s not clear-cut n the context of a country such as Australa, where low levels of domestc fertlty are generally not consdered a problem (page 29). The Australan Government has not specfed an optmal populaton level for a number of reasons. Chefly, there s no clear formula for a workable populaton polcy n a developed country wth low fertlty (page 45). These statements mply that only a country wth hgh fertlty needs to specfy an optmum populaton, and that, f fertlty s low, there s no need to specfy an optmum populaton or a populaton polcy. The offcal vew thus fals to take account of the fact that mmgraton can add sgnfcantly to populaton numbers. As Table 6 shows, the contnuaton of levels of annual net mgraton smlar to those experenced durng recent years and the current below-replacement fertlty rate produces an annual numercal ncrease equvalent to a much hgher fertlty rate. For example, an annual net mgraton of 80,000 combned wth the exstng total fertlty rate (TFR) of 1.8 chldren per woman, produces a numercal ncrease equvalent to a TFR of 2.4 chldren per women. An annual net mgraton of 160,000 produces a numercal ncrease equvalent to a TFR of 3.0 chldren per woman. Countres wth a TFR near 2.4 n 1993 nclude: Chna (2.2), Democratc People's Republc of Korea (2.4), Thaland (2.2), and Sr Lanka (2.5), whle countres wth a TFR near 3.0 n 1993 nclude ndonesa (2.9), and Brune Darussalam (3.0) (Unted Natons Economc and Socal Commsson for Asa and the Pacfc, 1993). f Australa's fertlty were above replacement level at an average or 2.4 to 3.0 per woman, ths would be of some real concern wth regard to populaton growth. Contrary to ths, current government polcy n Australa gnores the fact that the equvalent effect s produced from the combnaton of below-replacement fertlty and recent levels of mmgraton. 34

17 Chrstabel Young and Lncoln Day Table 6: Australa's atttude to populaton polcy mpled by the statements n Australa. Natonal Report on Populaton 1994, compared wth the deal response (n brackets) Stuaton Need for an optmum populaton target? Need for a populaton polcy? Low fertlty TFR = 1.8 No need No need Zero net mgraton Hgh fertlty Zero net mgraton (No need) Yes (Yes) (Yes) Yes (Yes) Low fertlty TFR = 1.8. No need Yes. Mantan mmgraton Moderate-hgh (Yes) (Yes. Lnk mmgraton to mmgraton the populaton target) 77ns represents a numercal ncrease each year equvalent to: TFR of 2.4 f annual net mgraton s 80,000* TFR of 3.0 f annual net mgraton s 160,000* * Accordng to the number of brths and the total fertlty rate (TFR) n The Natonal Report s lsted under Natonal Commttee (1994) n the references. Our vew, as ndcated n Table 6 n brackets, s that n Australa's present crcumstances, a populaton polcy and a populaton target are both desrable. n partcular, nstead of the exstng offcal polcy to mantan mmgraton, a more logcal approach would be to lnk the annual mmgraton ntake to the ultmate populaton target. Current practce also rases the queston: How can t be possble to specfy an annual mmgraton target wthout havng a populaton target? On page 33 of the Natonal Report we are told that the fgure 80,000 vsas for was arrved at because of the wdespread vew that t was 'about rght' not a very scentfc approach. 35

18 Populaton 2040 The stuaton of an mmgraton-orented department beng gven responsblty for populaton matters n Australa also results n an nconsstency n the assessment of the contrbuton to the populaton from mmgrants and from brths. The offcal vew seems to attrbute economc, socal and humantaran goals solely to mmgraton, and to overlook the socal and psychologcal value of chldren to famles and to the communty, and the potental skll output from educatng the chldren born n Australa. The Natonal Report devotes a great deal of space to a dscusson of servces and programs for mmgrants, but there s no menton of the status and drecton of school, trade and unversty tranng of our chldren; of the low status and ptfully low salares of school teachers n Australa; of the problem of youth unemployment; or of the lack of servces for dsadvantaged young people. The Natonal Report also shows a lack of consstency n the assessment of the lnk between populaton sze and the envronment compared wth the lnk between populaton sze and economc well-beng, by acceptng the questonable fndngs from economc models whch clam postve benefts of a larger populaton, whlst rejectng the more robust studes showng the deleterous effect of the growth of human populatons on the envronment. (Dovers et al, 1992; Douglas, 1977; Hopper and Coates, 1992, and the other chapters n ths publcaton). Optmum populaton and possble scenaros The optmum populaton s usually thought of n terms of total numbers, and that s both ts most mportant and most dffcult to defne characterstc. t s easer to be specfc about the other desrable demographc propertes of an optmum populaton: (a) contnung low age-specfc mortalty rates; (b) a gradual attanment of the populaton target, eventually producng a populaton whch s nether ncreasng nor decreasng; and (c) a regular age/sex structure. As for ts sze, ths would vary dependng on people's values and lfestyles relatve to: (a) rates of resource consumpton; (b) the uses made of the resources consumed; (c) patterns of settlement; and (d) the amenty value placed on, for example, access to open space and the enjoyment of nature. But whatever ts sze, to be an optmum, t would not be so great as to threaten the contnuaton of ecologcal sustanablty. Achevement of these optmum demographc condtons would, of course, be no guarantee of the 'good lfe'. For any partcular pool of resources, low mortalty, an unchangng age/sex structure, and a zero growth rate would only make the 'good lfe' more attanable; they would not, n themselves create t (Day, 1971). Because net mgraton s both the major potental source of populaton change and the component most under the control of government, t s nterestng to examne the logc behnd varous possble scenaros for selectng the future level 36

19 Chrstabel Young and Lncoln Day of net mgraton. As an example, we have used a populaton target of mllon n 2040, representng what mght be regarded as the low feasble range, as shown n Table 7. There are two man scenaros: (a) a constant level of annual net mgraton; or (b) an ncreasng level of annual net mgraton. The frst approach s smpler and has the advantage of a constant annual ntake. The second approach would generally requre that annual net mgraton was low over the next 20 years and then ncreasng. Compared wth a contnuous ntake whch produced the same ultmate populaton n 2041, the second approach would have the advantages of preventng hgh mmgraton whle there s an excess labour supply and of beng more effectve n retardng the ageng of the populaton. However, f the populaton target for 2040 was n the regon of mllon, the second approach would ntally requre levels of net mgraton below the feasble mnmum of 50,000 (for a populaton wth fertlty remanng at 10% below replacement). Let us brefly examne the logc behnd varous scenaros whch requre a low then ncreasng annual net mgraton (as gven n Table 7). The frst, n whch the annual level of net mgraton s ncreased solely because the populaton ncreases, so as to produce a constant rate of net mgraton, s llogcal. The second, n whch the annual level of net mgraton s ncreased when natural ncrease decreases, would have the advantage of a constant total ncrease; however, ths s equvalent to usng net mgraton (whch tends to be concentrated at the adult ages) to balance the ncreasng number of deaths (of mostly elderly people), and so s somewhat llogcal. The thrd, n whch net mgraton would be ncreased when the annual number of brths decreased s more logcal, because t would result n a constant annual number of new recruts to the populaton. ts fault s that net mgraton s not equvalent to brths because mmgrants arrve mostly at the adult workng ages. Consequently, the fourth opton s the most logcal n that annual net mgraton would be lnked to the hstorcal trend n brths, changes n the age structure and movement nto and from the workng ages. Thus, the hgh levels of mmgraton n the 1950s and 1960s were logcal because of the defct number of brths 20 to 30 years earler n the 1930s and early 1940s. Ths approach s also the most effectve n preventng a further excess supply of labour before 2011 and n offsettng the ageng of the populaton durng 2011 to

20 Populaton 2040 Table 7: Possble scenaros for determnng the level of annual net mgraton (NM) suffcent to acheve a populaton no greater than 24 mllon n 2041 Scenaro Advantages Dsadvantages Constant annual net mgraton (provded NRR remans constant) Suffcent constant annual net mgraton so that there s no declne n populaton before 2041, and no ncrease n next decade Suffcent constant annual net mgraton so that annual brths reman nearconstant Net mgraton (NM) remans constant (50,000 f NRR = 0.9) Constant annual no. of brths and hence constant no. of new recruts to workng ages n 20 years tme NM remans constant (50,000 f NRR = 0.9) Annual net mgraton low and then ncreasng Constant rate of net mgraton,.e. NM ncreases as populaton ncreases Relate net mgraton to natural ncrease (N),.e. NM ncreases as N decreases Relate net mgraton to annual number of brths,.e. NM ncreases as number of brths decreases Relate net mgraton to the hstorcal trend n brths, and to changes n the age structure, especally entres to and exts from the workng ages Constant annual total ncrease More effectve n retardng the ageng of the populaton. Constant annual number of new recruts to the populaton Constant balance between number of new entrants to the workng ages and the number of exts due to old age Would prevent hgh mmgraton whle there s an excess labour supply Most effectve n retardng the ageng of the populaton llogcal to ncrease NM solely because the populaton ncreases Equvalent to usng NM to balance the ncreasng number of deaths (of mostly elderly people) NM would have to be below feasble lmt before 2011 mmgrants are not equvalent to brths, as they add mostly to the workng ages More complex approach NM would have to be below feasble lmt before

21 Chrstabel Young and Lncoln Day Now let us turn to the thrd requrement of an optmum populaton, namely, a regular age structure. One requrement for ths s a relatvely constant number of brths. As shown n Fgures 6 and 7, the stuaton of a declnng number of brths when fertlty s 10% below replacement and there s no mgraton can be transformed nto a stuaton of constant annual brths f annual net mgraton s 50,000. Ths stuaton would also produce a regular age structure (see Fgure 8). A hgher level of annual net mgraton would produce an ncreasng numbers of brths, and therefore an rregular age structure (see Fgure 9). However, mmgraton cannot perform ths useful effect f fertlty s much lower than replacement (for example, a NRR below 0.8). Although a suffcent level of annual net mgraton can be found to brng about a constant number of brths, ths would be countered by the bulge n the age pyramd created by such a hgh level of net mgraton (see Fgures 10 (a) and (b)). Therefore, the necessary condtons for a regular age structure n the optmum populaton are: (a) a near constant number of brths; and (b) fertlty near replacement. 39

22 Populaton a 50 o Brths * Deaths Fgures shown refer to average annual brths and deaths durng each fve-year perod 1 ( > Fve-year perod begnnng n Source: Young (1990b) Fgure 6: Projected numbers of brths and deaths wth a net reproducton rate of 0.9 and zero net mgraton *' ^ ' 150- w -a E H Brths Deaths Fgures shown refer to average annual brths and deaths durng each fve-year perod ( Source: Young (1990b) Fve-year perod begnnng n Fgure 7: Projected numbers of brths and deaths wth a net reproducton rate of 0.9 and annual net mgraton of 50,000 40

23 Chrstabel Young and Lncoln Day ) mm cz : r_ r~ 3.00 MALES Populaton (%) 3.00 FEMALES Rate of annual net mgraton = 0.33% Fertlty reaches replacement level n 2022 Source: Day (1994) Fgure 8: Australa. Projected age structure of the populaton n 2042 accordng to medum levels of fertlty and net mgraton 1 MMM u M MALES Populaton (%) FEMALES Rate of annual net mgraton = 0.85% Fertlty reaches replacement level n 2002 and 10% above replacement n 2037 Source: Day (1994) Fgure 9: Australa. Projected age structure of the populaton n 2042 wth fertlty close to replacement and hgh net mgraton 41

24 Populaton 2040 K» HVM « Q * _C, FEMALES ] ' 9 86 resdent populaton and ther chldren bon s ) Post 1986 mgrants and ther chldren born snce arrval Source: Young (1989) Fgure 10(a): Australa. Projected age structure of the populaton n 2046 wth medum fertlty (net reproducton rate of 0.9) and medum annual net mgraton (of 50,000) 85* 80* J * * * Populaton f000) FEMALES 1986 resoew populaton and ther chldren bom snce 1986 QJ Pott 1986 mpanu and ther chldren bom snce arrval Source: Young (1989) Fgure 10(b): Australa. Projected age structure of the populaton n 2046 wth low fertlty (net reproducton rate of 0.7) and hgh annual net mgraton (of 150,000) 42

25 Chrstabel Young and Lncoln Day Further observatons on populaton growth Australa's populaton s hghly concentrated n a narrow coastal frnge. At the 1991 Census, 64% of Australa's populaton was lvng n the captal ctes, and 85% was lvng n urban centres (ABS, 1994c). Less than 50% of the populaton lved n the major urban areas n the pre-war years, but ths grew to 85% durng the 1970s, and has remaned near ths fgure ever snce (Young, 1992). Because a hgh proporton of mmgrants settle n the ctes and other major urban centres, mmgraton has accelerated the growth of ctes and the process of urbansaton. Much of any future populaton ncrease seems lkely to occur n the major urban areas, at least partly because the majorty of mmgrants settle n the ctes, and also because there s lttle scope for expanson of the farmng sector. An ncrease n the populaton to 23 mllon, for example, s lkely to result n 40% ncrease n the populaton of the urban areas, whle an ncrease to 37 mllon, as a result of hgher mmgraton, would have an even greater mpact on our urban lfestyle. No populaton can go on ncreasng forever; there are lmts physcal, ecologcal, and socal. Although these lmts can be extended by changng our pattern of use of the envronment and the pattern of behavour of ndvdual members of socety, there wll be a pont even wth the most judcous use of the envronment and the most prudent of human behavour beyond whch ncreases n populaton wll nevtably result n declnes n the qualty of lfe. t does not take much wt to see that ths has already happened n many places n Australa, not to menton elsewhere n the world. Fnally, t must be acknowledged that refugee mgraton s a specal case, because countres have an oblgaton to refugees that they do not have to other types of mgrants. Wth regard to refugees, a reasonable polcy for Australa would be to accord refugees prorty n any mmgraton program, but also to recognse that: (a) refugee status s necessarly a matter of degree; (b) there s only so much that a country lke Australa can do; (c) the natural rght to leave one's country s matched by no complementary rght to settle n whatever country one pleases; (d) resettlng them onto ts terrtory s only one of the means and seldom a very satsfactory one that a country has of helpng people; and (e) the most effectve acton on behalf of refugees s to prevent or amelorate the condtons that cause people to become refugees n the frst place (Day, 1991). 43

26 Populaton 2040 Concluson t s desrable that Australa establsh general demographc goals for tself. Ths wll: (a) (b) (c) provde assstance n establshng the ntermedate polces necessary for the attanment of desrable demographc condtons; provde a set of crtera for assessng progress towards achevement of the demographc goals that have been set; provde a framework for necessary acton and polcy formulaton n related non-demographc areas. The frst step towards decdng on an optmum populaton s be aware of the sze of the populaton n 2040 whch would result from feasble ranges of mortalty, fertlty and net mgraton, and to note that the potental mpact of mmgraton on the future sze of the populaton s consderably greater than that of mortalty or fertlty. The actual sze of the 'optmum' populaton s too much a functon of lfestyles and values to permt ts beng determned n more than a very general way. But, although t may not be possble to fnd unversal agreement about the exact populaton target, there should at least be some attempt to specfy a desrable feasble range for future populaton sze. n Australa, the scope for a government to change the level of fertlty or the level of mortalty s small compared wth the scope for changng the annual level of net mgraton. Moreover, whle the combnaton of brths and deaths and the age structure wll automatcally produce a declne n natural ncrease and progress towards lower levels of populaton ncrease, there are no foreseeable naturally occurrng brakes on mmgraton. mmgraton thus remans the major tool for reducng rates of populaton growth. A ratonal decson about long-term populaton goals for Australa wll requre a shft n offcal atttudes towards populaton growth and a recognton that mmgraton s not the sole source of socal and economc goals, but that brths, the other means of recrutment to the populaton, play an mportant socal, psychologcal and economc role n the development of the populaton. The thrd requrement of an optmum populaton, namely a regular age structure, can be met only f fertlty s near replacement and a moderate level of net mgraton prevals. mmgraton s nether a suffcent nor an endurng means by whch to prevent transton to an older age structure, nor can t prevent a dstorton n the age structure f fertlty falls very far below replacement. The natural ablty of a populaton, wth near replacement fertlty, to grow more slowly, whlst stll mantanng ts low age-specfc mortalty rates, and wth 44

27 Chrstabel Young and Lncoln Day just enough net mgraton to mantan a relatvely constant number of brths and a regular age structure, s a valuable and unque property of the Australan populaton at the present tme. t would be foolsh not to take advantage of ths fact. References Australan Bureau of Statstcs (varous years) Brths Australa, Cat. No , Canberra. Australan Bureau of Statstcs 1990, Projectons of the Populatons of the Australan States and Terrtores 1989 to 2031, Cat. No , Canberra. Australan Bureau of Statststcs 1993, Deaths Australa 1992, Cat. No , Canberra. Australan Bureau of Statstcs 1994a, Estmated Resdent Populaton by Sex and Age. States and Terrtores of Australa. Prelmnary ]une 1992 and June 1993, Cat. No , Canberra. Australan Bureau of Statstcs 1994b, Projectons of the Populatons of the Australan States and Terrtores 1993 to 2041, Cat. No , Canberra. Australan Bureau of Statstcs 1994c, Populaton Growth and Dstrbuton n Australa Census, Cat. No, , Canberra. Beaujot, R. 1989, "The role of mmgraton n changng demographc structures: The Canadan case', General Conference of the nternatonal Unon for the Scentfc Study of Populaton, Sesson 1-10, New Delh. Coale, AJ. 1987, 'Demographc effects of below-replacement fertlty and ther socal mplcatons', n Below-Replacement Fertlty n ndustral Countres, eds K.Davs et al, Supplement to Vol.12 Populaton and Development Revew, The Populaton Councl, New York, pp Day, L.H. 1971, 'Concernng the optmum level of populaton', n 7s There an Optmum Level of Populaton?, ed S. Fred Snger, New York, McGraw-Hll. Day. L.H. 1991, "The oblgaton to refugees', n mmgraton, Populaton and Sustanable Envronments: The Lmts to Australa's Growth, ed J.W. Smth, The Flnders Press of Flnders Unversty, Bedford Park. Day. L.H. 1992, The Future of Low-Brthrate Populatons, London and New York, Routledge. Day, L.H. 1994, 'Demographc projectons for the Stone Commttee', Demography Program, Research School of Socal Scences, Australan Natonal Unversty, Canberra. Denton, FT and Spencer, B.G. 1987, 'Changes n Canadan populaton and labour force: prospects and mplcatons', Canadan Studes n Populaton, vol. 14, no. 2, pp

28 Populaton 2040 Douglas, an 1977, 'Fronters of settlement n Australa-ffty years on', Australan Geographer, vol. 13, no. 5, pp Dovers, Stephen; Norton, Tony; Hughes, an and Day, Lncoln 1992, Populaton Growth and Australan Regonal Envronments, Canberra, Australan Government Publshng Servce. Hopper, S.D. and Coates, D.J. 1990, 'Conservaton of genetc resources n Australa's flora and fauna', Proceedngs of the Ecologcal Socety of Australa, vol. 16, pp Hugo, Graeme 1993, 'Recent trends n fertlty n Australa', People and Place, vol. 1, no. 2, pp Natonal Commttee 1994, Australa. Natonal Report on Populaton, for the Unted Natons nternatonal Conference on Populaton and Development Caro Text co-ordnated by the Natonal Commttee Secretarat (Bureau of mmgraton and Populaton Research and Department of mmgraton and Ethnc Affars), Canberra, Australan Government Publshng Servce. Unted Natons, Economc Commsson for Asa and the Pacfc 1993, ESCAP Populaton Data Sheet, ESCAP, Populaton Dvson, Bangkok. Young, CM. 1989, Australa's populaton: a long term vew, Current Affars Bulletn, vol. 65, pp Young, CM. 1990a, Australa's Ageng Populaton. Polcy Optons, Bureau of mmgraton Research, Canberra, Australan Government Publshng Servce. Young, CM. 1990b, Unpublshed projectons, updatng those n Young (1990a). Young, CM. 1992, 'Populaton growth and Australan ctes', Proceedngs of the Australasan Socety for Human Bology, vol. 5, pp