Age Bias in the Australian Welfare State (ANU 2013)

7/27/2019 Age Bias in the Australian Welfare State (ANU 2013)

1/16

5

Age Bias in the Australian

Welfare StateAlan Tapper, Alan Fenna and John Phillimore1

Abstract

This paper uses Australian Bureau o Statistics scal incidence gures to track

trends across the period 1984 to 2010 in one key aspect o the Australian welarestate whether welare policies have avoured the elderly at the expense o the

young. Our three main ndings are: that there has been a substantial shit over

this period in avour o the elderly; that this trend has accelerated rapidly in recent

years; and that as a result o this accelerated trend, elderly households today are on

average well o by comparison with younger households. We see little infuence o

party politics or ideology on the processes we are describing.

The lifecycle dimension of the Australianwelfare state

At the most general level, the welare state is in part about vertical equity the redistribution rom those who have more to those who have less. Butindividuals within any income or wealth bracket vary according to gender, ageand amily responsibilities, and social policy also takes an interest in these so-called horizontal categories. In this paper we are interested in the horizontaldimension o age. Our ocus is on liecycle stages. How does the Australian

welare state treat the elderly as compared with young adults and adults in mid-lie? We present an analysis o this key aspect o the Australian welare state,based on the evidence o Australian Bureau o Statistics (ABS) scal incidencesurveys. There are six surveys, taken at ve-yearly intervals, beginning in 1984and ending (most recently) in 200910. We aim to determine whether and, iso, to what extent spending and taxing have tended to avour the elderly anddrited away rom supporting the young across this 26-year period.

Leading Australian welare analysts today tend to see the welare state in termso its impact on both age and amily. In their 200102 scal incidence study,

Harding et al. (2006: 195) concluded that the impact o the welare state ...

1 John Curtin Institute o Public Policy, Curtin University; [email protected]


2/16

Agenda, Volume 20, Number 1, 2013

6

varies greatly by household type, with older Australians and sole parentsemerging as the largest gainers rom redistribution. They added that whilethere is substantial redistribution toward lower income couples, on averagecouples with children are not net gainers rom the taxes and benets consideredin our study. Cox (2003: 111) reached the same conclusion in his cross-nationalanalysis: In general, the welare state redistributes away rom childless couplesunder retirement age and couples with children towards single people, soleparents and the elderly. The extent o redistribution away rom couples withchildren is greater in large than in small welare states. In Australia, he noted,this redistributive eect is relatively small but the pattern is the same. Puttingaside the case o support or sole parents, on these views the Australian welarestate transers resources rom the relatively income-rich but asset-poor stages o

youth and mid-lie to the relatively income-poor but asset-rich stages o laterlie.

These analyses raise two questions. First, what is the relative magnitude o theseliecycle transers, when compared with vertical redistribution? Using datarom the early 1990s, Falkingham and Harding (1996, as summarised in OECD2008: 100) quantied this as ollows: in Australia, 38% o lietime benetsreceived by individuals, on average, were nanced through taxes they paidat another stage in their liecycle, and the remaining 62% o lietime benetsinvolved redistribution between rich and poor. We wont pursue that point

here. Secondly, how stable is this pattern o liecycle transers across time? Someanalysts contend that as welare states mature, policy avouritism drits rom theyoung to the elderly. The growing bias o the welare state towards the elderlyis a theme canvassed internationally (see, or example, OECD 1988; Thomson1996; Thomson 1999; Pierson 2006: 21221). In the Australian literature, theocus has been on the scal sustainability o current policies given concernsabout an ageing population (PC 2005; Harmer 2009; Henry 2009). Our purposehere is to analyse what has happened to age-related transers in Australia in thepast ew decades. Our hypothesis is that expressed by Thomson (1996; 1999):

net social support has shited away rom the young and towards the elderly.

Estimating nal income

Fiscal incidence analysis quanties the eects o government social expenditureand taxation on household incomes. In this analysis, social expenditure includescash expenditure on pensions, benet and allowances, but it also counts in-kindexpenditure on health care, education and housing these are estimated on

the basis o government spending. The taxation gures include both personalincome tax and consumption taxes such as the Goods and Service Tax wheneverthese can be ascribed to households. By taking into account both in-kind social


3/16

Age Bias in the Australian Welfare State

7

expenditures and taxation, scal incidence analysis provides a richer pictureo the welare state than conventional analysis, which tends to stop at the levelo direct social expenditures. Indirect social expenditures total 50 per centmore than direct social expenditures, and progressive income tax has a greaterredistributive eect than direct social expenditure (Fenna and Tapper 2012).Fiscal incidence analysis includes all levels o government, and is thereoresuperior to simply tracking Commonwealth budget trends as a means o gaugingpolicy trends.

The key results o this kind o analysis are the nal income and the net benetsor each household type. Final income is the net o private income, taxes onincome, the Medicare levy, amily tax benets, taxes on production, and social

expenditures in cash and in-kind.2

Net benet is the sum o social expenditureminus household taxes. The scal incidence approach makes it possible tocompare private incomes with nal incomes, with the dierence between thetwo being the redistributive impact governments have on household budgets.The ABS scal incidence gures include data on households classied by age ohousehold head. It is now possible to track household types classied by agegroup across 26 years. This permits us to examine age-related changes in welareredistribution across time, thus moving beyond static pictures o the welarestate at particular points in time.

All else being equal, policy intentions and actions should correlate with scalincidence evidence at the household level. But all else is rarely equal. Spendingincreases intended or particular categories may be swallowed up by increasedtaxes or unnoticed losses in other parts o the budget. Lynch (2006: 4) hasattempted to measure the age orientation o welare states using what she termsthe Elderly/Non-elderly Spending Ratio, dened as the ratio o direct socialexpenditures on the elderly (pensions and services or the elderly) to spendingon the non-elderly (unemployment benets, active labour market policy, amilyallowances and amily services), adjusted or the relative size o the elderly andnon-elderly populations. This is very useul as ar as it goes; however, thismeasure suers rom its ailure to include taxation. The strength o the scalincidence surveys is that they track both social expenditure and taxation onhouseholds. The sum o the two is the net benets o taxing and spending. Wethink this gives a more robust guide to policy trends and eects.

It is important to note that the scal incidence gures do not track actualhouseholds but household types.3 As their circumstances change, individual

2 All dollar gures in this paper have been adjusted or infation, using 2010 as the base year. In the tablesbelow, the gures shown are the mean or the sample o the household type. Taxation gures are not actual

reported gures but are imputed rom standard tax rates. For some cautions about using ABS gures to makecomparisons across time, see ABS Cat. No. 6537.0, 2001, 5355.3 Tracking actual households would require a panel study such as the Household, Income and LabourDynamics Australia (HILDA) survey. At present HILDA has just 10 years o data.


4/16


8

households move in and out o these categories. The scal incidence trendsindicate changes in policy outcomes, not changes in actual households. However,because the period covered by the scal incidence surveys includes 15 years oLabor government in Canberra (198496 and 200710) and 11 years o Coalitiongovernment (19962007), we can attempt to match up policy intentions withpolicy outcomes. How ar did Labor or the Coalition achieve what they setout as their policy aims? In this paper, we will be both analysing changes inthe ortunes o dierent age groups and amily types while also exploring thequestion o how policies have translated at the household level.

Policy 1983 to 2010

Labor, 19831996

The cornerstone o the Australian welare state is the age pension, untilrecently available to men at age 65 and women at age 60. The scheme has alwaysbeen non-contributory and means tested, with both income and assets tests.The Fraser government had removed the assets test in 1976, but the Laborgovernment restored it in 1985, signiying a desire to keep elderly entitlementsunder control. Since 1972, when the Whitlam government raised the single agepension rom 20 to 25 per cent o male average weekly earnings, it has stayed

close to the 25 per cent mark (Harmer 2009: Chart 4: 33). Labors commitmentto tight control o pension costs was expressed by ormer Finance MinisterPeter Walsh. Responding to those who say we have paid taxes all our lives andthereore we are entitled to a pension, he remarks:

Except or a brie period between 1976 and 1984, there has always beenan assets test, which prior to 1976 was much more rigorous than it is now.Todays aged may or may not have paid taxes and, i they did, they paidtaxes commensurate to unding a much more targeted pension, paid at amuch lower rate to a smaller proportion o a smaller aged population. By any measure, todays aged are treated generously more generouslythan they treated the aged when they were o prime working age (Walsh1995: 1123).

Labor did introduce one radical change: compulsory superannuation, intendedultimately to replace the age pension or most wage and salary earners.Employers were required to withhold a set percentage o employee wages orsalaries or accumulation in a superannuation und. The aim was a dual one: toimprove the adequacy o retirement incomes, and to reduce the long-run cost

o retirement incomes to the public purse, especially in light o demographictrends. This applied, however, to a uture generation o elderly, not to thosecurrently nearing retirement.


5/16


9

The Coalition, 19962007

In the retirement incomes system, the Coalition conrmed in law the principle

that the single age-pension rate was to be 25 per cent o male total averageweekly earnings. It was as willing to provide substantial though means tested

support to the elderly as to support amilies with children. It also acceptedthe principle o compulsory superannuation, though it declined to introduceurther increases beyond the nine per cent contribution level.

Labor 200710

The Rudd government was elected on a commitment to supporting working

amilies. Nevertheless, a major review o the nancial status o seniors wasinstigated, published as The Retirement Income System: report on strategic issues(Henry 2009). This led to a substantial increase in the single age pension in 2009.

Taxes, benets and age groups

In the next two sections we summarise trends at the household level, using agegroup data or the period 1984 to 200910. Is the welare state skewed towards

the elderly? And has there been a shit in welare priorities rom young to old?The scal incidence gures allow us to discover any skewing and to track thiskind o trend. The ABS breaks up its incomes data into age brackets. In Table 1we compare the net benets to each age group rom 1984 to 200910.

Table 1: Net benets by age group, 1984 to 200910 (in 2010 dollars per

household per week)

Ages 1984 198889 199394 199899 200304 200910Change 1984

to 200910

1524 -68 -127 3 47 -5 14 +82

2534 -113 -200 -50 -50 -149 -145 -32

3544 -21 -105 -13 18 -21 44 +65

4554 -129 -179 -131 -168 -171 -96 +33

5564 18 -19 78 58 -17 -7 -25

65+ 341 372 422 416 436 602 +261

All 14 -37 53 51 18 91 +77

Source: ABS Cat. No. 6537.0. Net benets is dened in the ABSs Explanatory notes as total benetsminus total taxes.

I we compare each age group with the trend or all households, two thingsstand out. First, there are large losses to two age groups: those aged 2534 (down


6/16


10

$109pw compared with the general trend) and those aged 5564 (down $102pw);and there are large gains to those aged 65 and over (up $184pw). Support hasswung sharply towards the end o lie.

In Figure 1, we compare nal incomes in these age groups across 26 years. Thegeneral trend (not shown in Figure 1) is fat or the rst decade, rises slightly inthe second, and rises sharply in the last six years; overall there is a 59 per centgain. The relevant question is whether the ve age groups rise or all relative tothat general trend. The youngest groups, 1524, 2534 and 3544, gain by 44per cent, 45 per cent and 47 per cent respectively. The middle age group, 4554,gains by 50 per cent. The 5564 age group remains rather fat in the rst twodecades, then leaps dramatically in the last period, gaining 75 per cent overall.

Similarly, the elderly group, 65 and over, gains in the last ve years, rising 98per cent overall. Thus, the older groups gain relative to the general trend, whileyounger households all relatively. The swing avouring the elderly is largely aeature o the 2000s.

Figure 1: Trends in nal incomes by age group, 1984 to 200910

Source: ABS Cat. No. 6537.0.

How should we interpret the trends in the age group data? In Table 2 we look at

the break-up across time of taxes and expenditures for three household types. Weinclude couples with dependent children, since they are one useful benchmark

against which to compare the fortunes of elderly households. For a broader and

more basic benchmark, we also compare the elderly with all households.


7/16


11Table2:Taxesandbenetsforthree

householdtypes,1984to

200910(in2010dollarsperhouseholdperweek)

1984

198889

199394

199899

200304

200910

Change1984

to200910

Directsocial

assistance

Coupleswithchildren

78

64

108

110

132

135

+57

65andover

286

271

289

259

296

349

+63

All

households

137

120

151

148

164

177

+40

Educationbenefts

Coupleswithchildren

189

189

194

245

271

308

+119

65andover

8

7

5

3

3

9

+1

All

households

99

97

89

100

105

117

+18

Health

benefts

Coupleswithchildren

102

80

103

129

146

182

+80

65andover

113

159

149

192

231

316

+203

All

households

92

114

107

130

135

181

+89

Otherwelfare

benefts

Coupleswithchildren

10

17

22

42

52

64

+54

65andover

42

49

69

64

57

80

+38

All

households

19

25

33

42

42

46

+27


8/16


12

1984

198889

199394

199899

200304

200910

Change1984

to200910

Allb

enefts

Coupleswithchildren

386

354

431

527

602

691

+305

65andover

454

490

518

523

593

762

+308

All

households

354

358

380

421

451

534

+180

Taxesonincome

Coupleswithchildren

299

325

290

358

391

447

+148

65andover

66

60

47

42

49

40

-26

All

households

241

242

214

250

256

260

+22

Indirecttaxe

s

Coupleswithchildren

106

144

114

140

234

242

+136

65andover

47

58

47

64

108

120

+73

All

households

99

124

98

116

177

183

+86

All

taxes

Coupleswithchildren

405

470

403

498

625

689

+284

65andover

113

118

94

106

157

160

+47

All

households

340

366

312

366

433

443

+108

Netbenefts

Coupleswithchildren

-19

-116

28

30

-22

2

+21

65andover

341

372

422

416

436

602

+261

All

households

14

-8

68

55

18

91

+77

Source:ABSCat.No.6537.0.Educationbeneftsar

edefnedintheABSsGlossaryassocialtransersinkindderivedrom

governmentexpensesrelatingtothe

provisionosc

hool,tertiaryandothereducation.

Healthbeneftsaredefnedassocialtransersinkindderivedromgovernmentexpensesrelatingtoacute

careinstitution

s,communityhealthservices,pharm

aceuticalsandotherhealthbenefts.Otherwelarebeneftsincludess

ocialtransersinkindderivedrom

governmentex

pensesrelatingtotheprovisionogoodsandservicestospecifcpopulationgroupswithspecialneeds.Itincludesexpenditureonchildcareservices

(includingchildcarebeneftsubsidies),servicesor

theaged,servicesorpeoplewithad

isability,etc.


9/16


13

Table 2 shows how changes in taxation have played an important part in changesin net social support. Income tax has both grown and shited target. In 200910,those 65 and over paid $40pw in income tax, less than was paid in 1984; whilea couple with dependent children paid $447pw, 50 per cent more than in 1984.For all households, income taxes grew by only eight per cent in that period.Indirect taxes have also grown and shited target in a similar manner. They havegrown or the elderly, but not as much as or all households, and markedly lessthan or couple amilies. Overall, taxes on the elderly grew by 42 per cent, oncouple amilies by 70 per cent, and on all households by 30 per cent. Taxationchanges have disadvantaged couple amilies. The elderly have also increasedtheir share o taxation, but by a considerably lesser raction.

On the benets side, we can see that both the elderly and couple amilies gainedvery considerably when compared with all households. The biggest single actorhere was health care expenditures, which have both grown overall and swung inavour o the elderly. In 200910, those 65 and over received on average $316pwin health benets, while the average or all households was $181pw. Comparedwith 1984, the share or the elderly nearly tripled, while or all households theincrease was 100 per cent (or couple amilies it was 80 per cent).

It is worth noting that, as Table 3 shows, changes in household size play littlepart in these trends. Surprisingly, average household size has not changed in

this period or couples with children. Thus, there is no distortion in the trendsin social support caused by declining household numbers (ewer childrengetting less in total education support or example). On the other hand elderlyhouseholds have contracted a little, so analysed on aper capita basis their gainsrelative to couple amilies are to that extent greater than they appear in Table4. However, the average size o all households has also contracted by a similaramount, so on that comparison there has been no per capita gain to elderlyhouseholds.

Table 3: Household size, 1984 and 200910

1984 200910

Couples with children 4.1 4.1

65 and over 1.7 1.5

All households 2.8 2.6


Taking taxation and expenditure together, we can see rom Table 2 that theelderly have gained very strongly relative to all households, and even more

strongly relative to couple amilies. For our purposes, the comparison with allhouseholds is more important than that with couple amilies. All households


10/16


14

can be taken as a stable benchmark. Looking at the gains in net benets toelderly households relative to the trends or all households, what stands out isthe massive leap in support or the elderly between 200304 and 200910. Thisis much greater than the gains in the previous 20 years (up by $166pw in sixyears, as against $95pw in 20 years).

The actors operating here are varied, as Table 4 shows. While taxes stayedsteady, social expenditures were growing substantially. Overall benets perelderly household grew by 29 per cent. The largest growth was in healthexpenditure, up by 37 per cent.

Table 4: Taxes and benets to households 65 and over, 200304 to

200910 (2010 dollars per household per week)

200304 200910 Change ($) Change %

Direct social assistance 296 349 +53 +18

Age pension 227 253

Veterans pension 48 35

Disability support 7 15

Other pensions and allowances 8 36

Health benets 231 317 +86 +37

Acute care institutions 114 156

Community health services 56 68Pharmaceuticals 51 66

Private Health Insurance Rebate* 11

Other health benets 10 14

Other welfare benets 57 80 +23 +40

All benets 593 764 +171 +29

Taxes on income 49 40 -9 -18

Indirect taxes 108 120 +12 +11

All taxes 157 160 +3 +2

Net benets 436 602 +166 +38

* Previously included in Other health benets.


Age pension increases, though substantial, were not a major driver o theseexpenditure increases. In 2009 the Rudd government introduced an 18 per centincrease in the standard (single person) age pension and a six per cent increasein couples pensions.4 A urther increase ollowed in early 2010. Large though

4 Guide to Social Security Law, FAHCSIA website, 5.2.2.10 Rates o Pension July 1909 to Present Date.


11/16


15

this increase was, constant dollar age pensions per elderly household rose byonly 11.4 per cent between 200304 and 200910. The explanation is that agepension coverage was diminishing, as is shown in Table 5.

Table 5: Dependency on income support, 1984 to 200910 (Per cent

principally dependent on government pensions and allowances)

1984 198889 199394 199899 200304 200910

Persons 65 and over 72 69 72 67 69 61

All households 26.2 24.4 29.5 27.2 27.4 22.6


Figure 2 shows the overall history o taxes and benets or elderly households.What stands out is the recent surge in net benets, which took place mostlyunder Coalition government. Critics might argue that in this period the longAustralian tradition o relative restraint on expenditure on the elderly, underboth Labor and Coalition, was here abandoned.

Figure 2: Trends in taxes and benets for households aged 65 and over,

1984 to 200910



12/16


16

Incomes, assets and living standards

We might suppose that increasing support or the elderly is an expression oincreased recognition o need. To test this claim we need to be able to rankneediness. This can be partly done in terms o equivalent nal incomes,which are presented in the 200304 and 200910 ABS scal incidence studies.Equivalent nal incomes are nal incomes adjusted by an equivalence scale totake account o variations in household size. The resulting gure is an economicindex and not a dollar comparison. Table 6 shows the equivalent nal incomeso the elderly and all households. Note that this table exaggerates the increasein EFI between the two surveys, because here the EFI or 200304 has not beenadjusted by the CPI.5 The point o the comparison is not the relative changebetween 200304 and 200910 within each group, but the gains and losses othe dierent groups relative to each other in this period.

Table 6: Equivalent nal incomes, 200304 and 200910

200304 200910 Change 200304 to 200910

Persons 6574 489 939 +92%

Persons 75 and over 549 1006 +83%

Couples aged 65 and over 515 981 +90%

Singles aged 65 and over 475 900 +89%

All households 612 1012 +65%

Source: ABS Cat. No. 6537.0. Equivalent nal income is dened in the ABSs Glossary as householdincome adjusted by an equivalence scale. For a lone person household it is equal to household income. Fora household comprising more than one person, it is an indicator o the household income that would needto be received by a lone person household to enjoy the same level o economic wellbeing as the householdin question.

Table 6 shows, rather surprisingly, that older elderly (75 and over) are on averagebetter o than younger elderly (65 to 74). Couples are notably better o than

singles. But the most interesting comparison is that with all households. Rapidgains to the elderly in this period have brought them close to the EFI or allhouseholds. The gap in 200304 stood at about 21 per cent (based on an estimatethat the EFI or all elderly households was about 505). In 200910 it had allento only about ve or six per cent (estimating that the elderly EFI was about 960).

O course equivalent nal income is a measure o income adequacy, and takesno account o assets. Fortunately, the 200910 scal incidence study gives us aglimpse o the distribution o assets across the age groups. As Table 7 shows,wealthier households are older households. Net worth peaks at around age

5 Table 1 o the 2009-10 survey shows that EFIs rose by 27 per cent rom 2003-04 to 2009-10, not 65 percent as suggested here.


13/16


17

60. A sharper picture is obtained i we take household size into account usingequivalence scales. Here we have used the square root o household size (amethod that approximates quite closely to the OECD modied scale used bythe ABS to calculate equivalent nal incomes). The resulting equivalent networth indicates that even households aged 75 and over are one-third better othan the mean or all households, while households in the 6574 age group are60 per cent better o than the mean.

Table 7: Household net worth by age group, 200910

1524 2534 3544 4554 5564 6574 75+ All

households

Net worth

($2010)

76 460 256 702 553 547 989 253 1 068 851 950 959 764 561 729 442

Household

size

2.4 2.5 3.4 3.0 2.2 1.8 1.6 2.6

Equivalent

net worth

49 329 162 469 300 840 571 823 722 197 709 671 604 396 453 070


I we could combine equivalent income measures with the net-worth data, wecould arrive at a plausible assessment o relative living standards across the agespectrum. However, the scal incidence surveys do not report any integrated

measure o this sort. We can only guess at what the result might be. Giventhat equivalent nal incomes or the elderly are now close to the average or allhouseholds, and that the net-worth distribution is skewed in avour o olderhouseholds, we can reasonably iner that an integrated measure would showthat households headed by persons 65 and over are better o than the averageor all households under that age.

I all this is right, the Australian system o social transers to the elderly ismuch more than a saety net. Viewed in liecycle terms, it shits resources romthe income-rich but asset-poor stages o lie to the asset-rich but income-poor

stage. Viewed in terms o the vertical dimension, it is a system o upwardsredistribution rom the less well o to the better o.

Conclusion

In this paper we have been using a very broad denition o the welare state,seeing it as including standard cash transers (pensions and benets); in-kindexpenditures on health, education and housing; and taxation (direct and

indirect). Fiscal incidence analysis enables us to track social spending andtaxing at the household level. We have examined how the Australian welare


14/16


18

state, thus dened, has served various age groups through the 26 years rom1984 to 2010. Being ocused on households, the analysis is thereby independento demographic trends.

Our main ndings in this paper are threeold. The rst is that in this period theAustralian welare state has shited ocus in avour o the elderly. Substantialgains have been made by the elderly at the expense o those in the 25 to 54 agegroups.

Secondly, the pace o this trend has changed. From 1984 to 200304 there was aslow but steady drit towards increased support or the elderly. Ater 200304it accelerated. Gains to the elderly in this last period are driven by increasedbenets in health care, pensions and other welare support.

Thirdly, in part as a result o this sharp upturn, elderly households are todayalmost as well o as the average or all households when measured in terms oequivalent nal incomes. They are clearly better o than average householdswhen measured using equivalent net worth. The notion o a welare system thatredistributes upwards across age groups thus may be close to reality.

In response to our rst two contentions, one might reply that simply measuringtrends using the CPI to translate past gures into 2010 dollars is misleading.Two arguments might be given. One is that purchasing power parity is a better

indicator o changes in dollar values over time. This may be so; it is a line worthpursuing. But it wont explain why one group or category gains relative to othergroups or categories. All that the argument can show is that the relative gainsand losses are not best measured using the CPI. A second argument is that risinggeneral wealth should especially benet the elderly, because although theypaid relatively little taxation in earlier years, that little was a relatively largesacrice at that time when general standards were lower. Thomson argues thatwe should use some indicator o average male wages as an index, rather than theCPI, to take account o this (Thomson 1996: 167). Again, this is a line worth

pursuing. It may modiy our conclusions.In this study we have been tracking categories o household, not actual socialgroupings. However, when trends in social support move in tandem with theageing o certain cohorts, the eect is likely to be a avouritism that raisesquestions o intergenerational equity. To explore this urther would require acohort study, and we have not attempted that here.

Our three ndings require interpretation. What might explain them? It is noteasy to discern any party-political or ideological infuence on these trends. The

general drit towards avouring the elderly ran a slow and smooth course up to2004, under 12 years o Labor and eight years o the Coalition. Nearly our othe six years ater 2004 when this tendency suddenly accelerated were under


15/16


19

the Howard government but two were under the Rudd Labor government. Thepolitical explanation or this upturn, i there is one, is unclear. The main age-related policy innovation o these 26 years was the introduction o compulsorysuperannuation, accepted by all the main parties as a strategy or controllingage-related scal demands. But just when that policy seemed to be having someeect, elderly expenditure accelerated. Most likely gains to the elderly are notthe consequence o deliberate policy. They probably arise rom incrementaladjustments, no one o which is signicant but which cumulatively producesubstantial net eects.

References

ABS Cat. No. 6537.0, 1987, 1984 Household Expenditure Survey: Eects oGovernment Benets and Taxes on Household Income, Australia, Canberra.

, 1992, 198889 Household Expenditure Survey: The Eects oGovernment Benets and Taxes on Household Income, Australia, Canberra.

, 1996, The Eects o Government Benets and Taxes on HouseholdIncome: 199394 Household Expenditure Survey, Australia, Canberra.

, 2001, Government Benets, Taxes and Household Income, Australia,199899, Canberra.



Cox, J. 2003, The Distributional Analysis o the Welare State, Agenda 10(2),99112.

FAHCSIA, Guide to Social Security Law. Australian Government. http://guidesacts.ahcsia.gov.au/guides_acts/ssg/ssg-rn.html

Falkingham, J. and Harding, A. 1996, Poverty Alleviation versus SocialInsurance Systems: A Comparison o Lietime Redistribution, NATSEMDiscussion Paper No. 12, NATSEM, University o Canberra.

Fenna, A. and Tapper A. 2012, The Australian Welare State and theNeoliberalism Thesis, Australian Journal o Political Science 47(2), 15572.


16/16


20

Harding, A., Lloyd, R. and Warren, N. 2006, The Distributional Eects oGovernment Spending and Taxation, in D. B. Papadimtriou (ed.), TheDistributional Eects o Government Spending and Taxation, PalgraveMacmillan, Basingstoke, 176201.

Harmer, J. 2009. Pension Review Report. Canberra. http://www.ahcsia.gov.au/about-ahcsia/publications-articles/corporate-publications/budget-and-additional-estimates-statements/pension-review-report

Henry, K. 2009, The Retirement Income System: report on strategic issues.

Canberra: Department of the Treasury. http://www.taxreview.treasury.gov.au/

Lynch, J. 2006, Age in the Welare State: The Origins o Spending on Pensioners,

Workers, and Children, Cambridge University Press, Cambridge.

OECD 1988, Ageing Populations: The Social Policy Implications, OECD, Paris.

, 2008, Growing Unequal: Income Distribution and Poverty in OECDCountries, OECD, Paris.

Pierson, C. 2006, Beyond the Welare State? The New Political Economy oWelare, Polity Press, Cambridge.

Productivity Commission 2005, 'Economic Implications o an Ageing Australia',

Research Report, Canberra: Productivity Commission.

Thomson, D. 1996, Selsh Generations? How Welare States Grow Old, WhiteHorse Press, London.

, 1999, How History is Failing our Families: Public Policy and the Fall inFamily Fortunes since 1970, Family Matters 52: 1218.

Walsh, P. 1995, Conessions o a Failed Finance Minister, Random House, Sydney.

Documents

Age Bias in the Australian Welfare State (ANU 2013)