Fiscal Policy and Public Sector Saving: Historical Trends

The public sector underlying fiscal deficit, which had averaged around 2 percent of GDP during the 1960s and early 1970s (with the general government position in a surplus of ½ percent of GDP and the public trading enterprise sector in a deficit of 2½ percent of GDP) increased to an average of 4½ percent of GDP over the next decade.³ Although the deficit has since narrowed, it has remained above these earlier levels. Public sector saving has declined sharply, from an average of 8 percent of GDP in the 1960s to only 2½ percent of GDP so far in the 1990s, although there has been a good deal of cyclicality around this decline, with sharp falls in the recessions of the early 1980s and 1990s, but a strong increase in the late 1980s.

The deterioration in the public sector fiscal position is attributable to the general government sector, particularly the Commonwealth general government sector. The Commonwealth’s underlying budget position, which had been in surplus throughout the 1960s and early 1970s, deteriorated sharply in the mid-1970s (Figure 4.3). Since then, there has only been one period of sustained budget surplus, between 1987/88-1990/91, with a peak surplus of 1¾ percent of GDP in 1989/90. Significant deficits occurred in the downturns of the early 1980s and early 1990s, as the automatic fiscal stabilizers took hold, and the government actively used fiscal policy as a countercyclical policy instrument.

View Full Size

Commonwealth General Government Finances

(In percent of GDP, year ending June)

Sources: Australian Bureau of Statistics; Government Financial Estimates; and Commonwealth of Australia, Budget Statement, 1997/98.

• Download Figure
• Download figure as PowerPoint slide

Two main developments have underpinned the deterioration in the Commonwealth budget position: the significant expansion of expenditure in the mid-1970s and early 1980s as the government expanded its role in the economy; and the deterioration in the tax base from the mid-1980s caused by structural developments and discretionary policy decisions.

The Commonwealth government’s underlying expenditure rose from around 20 percent of GDP in 1973/74 to 26 percent of GDP two years later, and reached nearly 30 percent of GDP in the mid-1980s. Spending on education, health, and transfers to the states all increased noticeably, but the most important factor was the expansion of social welfare spending (Figure 4.4, top panel) with a move toward greater universalism in payments (for example, the means test on the age pension was gradually eased, and then abolished for those over 75 in 1973 and for those aged 70–74 in 1975).

View Full Size

Commonwealth Government Social Expenditure

(In percent of GDP, year ending June)

Source: Commonwealth of Australia, various budget statements.

• Download Figure
• Download figure as PowerPoint slide

Following the sharp deterioration in the Commonwealth’s budget position in the early 1980s, the government took steps to impart a more medium-term focus to fiscal policy, and a period of fiscal consolidation followed, based on expenditure savings in the areas of social security and welfare, transfers to state governments, and defense. The government made changes to the social security system, including better targeting of welfare payments through the greater use of income and asset tests, although several new income support programs were also introduced. These new programs have generally followed an “active” rather than a “passive” approach, and have sought to target those in need, while also increasing the real value of the payments made.⁴ Together with favorable economic developments (the unemployment rate fell from more than 10 percent in 1983 to less than 6 percent by late 1989), expenditure on social security and welfare declined in relation to GDP.

Underlying expenditure has risen again during the 1990s, and stood at just under 26½ percent of GDP in 1996/97. While this remains below the peaks of the early 1980s, Commonwealth own-purpose expenditure (i.e., excluding transfers to other levels of government) peaked as a percentage of GDP in 1995/96 (above the cyclical peak in the 1980s), indicating that much of the restraint in expenditure has been due to reductions in transfers. Spending on social security and welfare has again risen strongly. This is partly due to the increase in unemployment, but the Commonwealth has also introduced new discretionary spending measures (Figure 4.4, bottom panel), with the most significant growth occurring in assistance to families with children.

During the 1970s and the first half of the 1980s, revenues increased as a share of GDP, boosted by strong growth, by the high rate of inflation interacting with the nonneutralities with respect to inflation in the Australian tax system,⁵ and by new revenue measures, such as taxes on capital gains and noncash renumeration of employees. Since the mid-1980s, the revenue base has undergone some important changes, and the revenue-to-GDP ratio has fallen to around 25 percent of GDP. A series of personal income tax cuts specified by the Accord agreements in 1987/88 and introduced from 1989/90 to 1993/94 resulted in a decline in individual tax collections, which by 1995/96 were some 2 percent of GDP lower than in 1986/87. Structural change has also affected the tax base, with revenue falling as a result of tariff cuts and changes in the funding of state governments.⁶ The indirect tax base has also come under pressure due to changing consumption habits toward services, although revenues have broadly been maintained as a proportion of GDP by raising the tax rate on the remaining base.⁷

At the state general government level, the underlying deficit improved until the late 1980s, partly as a result of declining investment expenditure. The situation deteriorated, however, during 1988/89-1991/92, and the underlying budget recorded a deficit of 1 percent of GDP in 1991/92 compared to broad balance three years earlier (Figure 4.5). Revenue growth weakened markedly with the recession and the collapse of the commercial real estate market. Expenditures continued to grow and were further boosted by large budgetary payments associated with the financial difficulties of some state-government-owned or backed financial institutions. The deterioration in the fiscal position led the international credit rating agencies to downgrade the debt of several state governments.⁸

View Full Size

State and Local General Government Finances

(In percent of GDP, year ending June)

Source: Australian Bureau of Statistics, Government Financial Estimates.

• Download Figure
• Download figure as PowerPoint slide

Fiscal Consolidation and Policy Reform at the Commonwealth Level

The Commonwealth government has slowly withdrawn the fiscal stimulus it imparted to the economy in the early 1990s, and it has made progress in reducing the Commonwealth underlying budget deficit since 1992/93. By 1995/96 this consolidation had been slower than desirable, however, and Commonwealth government saving remained at a low level. Furthermore, with the budget remaining in deficit at a fairly advanced stage of the cycle, there had been little progress in reducing the outstanding stock of Commonwealth debt.

While Australia’s level of public debt is not high by international standards, the government believes that a failure to reduce the Commonwealth’s debt may limit its capacity to support periods of economic weakness in the future. To bring about a more rapid improvement in the budget position, the government announced in the 1996/97 budget a fiscal consolidation strategy that sought to move the Commonwealth budget sector from an underlying deficit of 2 percent of GDP in 1995/96 to a surplus in 1998/99. This consolidation is largely based on structural expenditure reductions in labor market programs, transfers to state governments, education (through greater cost recovery), and departmental running costs, including a significant reduction in public sector employment.

The underlying deficit declined to less than 1 percent of GDP in 1996/97, broadly as budgeted, and the 1998/99 budget estimates an underlying surplus of ½ percent of GDP, with surpluses projected to rise to 2 percent of GDP by 2001/02. The government is also seeking further savings by evaluating its outlay and tax expenditure programs, reviewing its involvement in some activities, and introducing competitive tendering and contracting for the delivery of some of its services.

The failure to withdraw the fiscal stimulus in the early 1990s as rapidly as originally envisaged led policymakers to focus on institutional reforms that could instill more discipline on the budgetary process by improving the transparency of fiscal reporting and by increasing the focus on the medium-term implications of fiscal policy action. Work in this direction began in the mid-1980s when several reforms were made to the budget process. The most significant of these was the introduction of rolling multi-year forward estimates that present the expected path of the budget under the assumption of no new policies over a three-year horizon. Furthermore, the reform set the starting point for budgeted expenditure in any year at the level included in the previous forward estimates for that year, and it required a reconciliation of the budget with the forward estimates. The distinction made in this reconciliation between policy and parameter variations presents the extent of the two influences on the fiscal position. Other reform measures taken to improve transparency included the publication of (1) the National Fiscal Outlook, which presents a consolidated picture of Commonwealth and states’ fiscal position, along with their fiscal strategies; (2) information on unfunded employee entitlements for the Commonwealth and the states; (3) details of annual tax expenditures; and (4) the underlying deficit measure, now the main focus of policy, which excludes from the headline deficit measure transactions that involve the transfer or exchange of a financial asset.

In a major step to reform the institutional framework for fiscal policy, parliament in early 1998 enacted the Charter of Budget Honesty. The Charter requires greater transparency and accountability in fiscal policy to improve the discipline of policymakers, which should help to strengthen the medium-term focus of policy (Box 4.1). Such fiscal responsibility legislation provides policymakers with an incentive to pursue greater fiscal discipline and addresses the inherent biases in policy making that make deficit reduction difficult. As such, it can play an important role in helping achieve prudent fiscal outcomes.

Different approaches can be taken to fiscal responsibility legislation, and there is generally a trade-off between providing policymakers with some degree of flexibility to conduct policy and establishing the credibility of a binding constraint. Fiscal responsibility legislation can establish certain reporting standards, set specific fiscal targets, or both. Adopting reporting standards would enable a more flexible fiscal policy response to changes in economic conditions, while the setting of fiscal targets may provide, in theory, more certainty that fiscal policy will follow a particular course. The Charter of Budget Honesty is similar to New Zealand’s Fiscal Responsibility Act, and the institutional reforms undertaken in both countries are at the forefront of policy efforts in this area. Both impose fiscal discipline by setting out broadly similar principles for legislated fiscal reporting requirements and responsible fiscal management. Neither the Charter nor the Fiscal Responsibility Act prescribes quantified fiscal targets, although, of course, the government is free to specify these as part of its announced fiscal strategy. In this regard, the Australian government has announced its intention of maintaining budget surpluses over the forward estimates period while economic growth prospects remain sound. In addition, the government has indicated that it intends to halve the ratio of Commonwealth general government net debt to GDP ratio from 20 percent of GDP in 1995/96 to 10 percent of GDP in 2000/01. The government has also announced its intention to implement accrual budgeting from 1999/00. The Charter of Budget Honesty has been drafted to accommodate this change.

Fiscal Reform at the State Government Level

The deterioration in the fiscal position of the states in the late 1980s and growing interstate and international competition prompted all the state governments to introduce some form of announced fiscal strategy.⁹ For example, New South Wales’ fiscal strategy has the medium-term objective of reducing net debt to sustainable levels and the long-term objective of eliminating general government net debt. This is to be achieved by restraint in outlays, better asset management, and improved service delivery, and by phasing in the full funding of accruing superannuation liabilities. In 1996, New South Wales introduced the General Government Debt Elimination Act, which sets out mandatory fiscal reporting requirements and nonmandatory short-, medium- and long-term fiscal targets. The Act also requires that government reporting be on an accrual basis, in line with generally accepted accounting standards (this has been the practice for several years). In Victoria, the long-term budget objectives set were to reduce state debt to a level consistent with a AAA credit rating (Standard and Poor’s upgraded Victoria from AA+ to AAA in April 1998); reduce Victoria’s tax burden closer to the average of the other states; deliver high-quality government services at least cost; and invest in infrastructure to contribute to a more productive economy.

The states have implemented a variety of measures to achieve their objectives, but they have generally focused on reforming public trading enterprises (discussed in Chapter 7); improving the efficiency of government departments; and addressing existing superannuation liabilities.

Within these strategies, some states have also fundamentally rethought the appropriate role of government. In areas where it was felt that there was no longer an adequate basis (on social equity or economic efficiency grounds) for public service provision or that the public sector was not in the best position to provide the service, privatization or contracting out have been used to reduce the role of the government. In areas that have remained under public ownership, governments have made efforts to improve the efficiency of service provision. Departments draw up business plans that specify which services they provide, the purpose of the services, and the government objectives they intend to satisfy. Performance measures set targets for the quantity, quality, and timely provision of output.

Annual productivity saving targets have been set for government departments and agencies to reduce recurrent expenditures. These have usually been of the order of 1–2½ percent of recurrent expenditure per annum. Substantial reductions in public sector employment have been made to reduce ongoing wage and salary costs. The majority of redundancies have been through voluntary packages, but these initially added significantly to expenditure in 1992/93 and 1993/94. Competitive tendering and contracting out have been introduced for an increasing share of contestable public sector services. This can improve the quality of the service provided, and produce substantial cost savings.

States have started to address the buildup of unfunded public sector superannuation liabilities and related outlays. The buildup in unfunded liabilities reflected the generous benefits provided under many state public sector superannuation schemes and the expansion of state public sector employment during the 1970s and 1980s. Most governments did not make explicit provision for the future payment of these liabilities, and by the early 1990s unfunded superannuation liabilities were in excess of 10 percent of GDP, and annual superannuation-related outlays accounted for more than 4 percent of total outlays. To address this problem, existing schemes were closed to new members, and less generous schemes were introduced, funded on an accrual basis.

The reforms have resulted in a significant improvement in the fiscal position of the states. The general government sector has recorded surpluses since 1994/95 following an improvement in fiscal performance since 1991/92. After peaking at more than 15 percent of GDP in the early 1990s, current expenditure declined to 13 percent of GDP in 1996/97. The revenue to GDP ratio has remained fairly constant as the growth in own-source revenue has offset the decline in transfers from the Commonwealth. The underlying surplus rose to ¾ percent of GDP in 1996/97, and recent official projections suggest the sector will achieve small surpluses over the medium term. The underlying budget surpluses, together with the significant proceeds from asset sales, have resulted in the ratio of net state debt to GDP declining from 20 percent in 1991/92 to less than 10 percent in 1996/97. The situations of different states vary widely, however, with net debt ratios ranging from –3 percent of Gross State Product in Queensland to 30 percent in Tasmania in 1995/96.

Private Saving

Private saving has fallen from an average of 16 percent of GDP in the 1960s (and 18½ percent of GDP in the 1970s) to only 14 percent of GDP so far in the 1990s. All of this decline is due to a drop in household saving, which has fallen from a peak of almost 15 percent of GDP in the mid-1970s to only 8½ percent of GDP in 1996/97. This decline has been greater than in most other industrial countries and has left Australia with one of the lowest household saving rates in the OECD.¹⁰ While fluctuating with the economic cycle, corporate saving—that is, the retained earnings of corporate enterprises—has averaged about 6 percent of GDP.

What Determines Private Saving?

A number of recent studies have investigated the determinants of private and household saving, both in an international context and for individual countries. These studies have generally found the following factors to be important in influencing saving behavior.

Public Saving

A direct link between government saving and private saving is often hypothesized. An increase in the government deficit due to a reduction in taxes may be offset by an increase in private sector saving, as higher tax liabilities in the future are anticipated (the well-known Ricardian equivalence argument). Most studies have found that while such an offset exists, it is less than unity (commonly around one-half) as implied by full Ricardian equivalence. However, Australian studies (see Blundell-Wignall and Stevens, 1992; and Edey and Britten-Jones, 1990) have suggested that there is little evidence of public saving affecting private saving.

Income Growth

While theoretically and empirically the impact of growth on saving is unclear, saving and growth are highly correlated over long time horizons, as well as for many regions and stages of development, with higher rates of growth being associated with higher saving (see Schmidt-Hebbel and others, 1996).¹¹ In Australia, real income growth declined from an average of 4¾ percent per annum in the 1959–74 period to 3 percent per annum between 1975 and 1996. The slowdown in per capita growth has been even more marked over this period, with Australia having had one of the lowest per capita growth rates in the OECD in the 1970s and 1980s (see Chapter 2). This growth slowdown is likely to have contributed to the decline in saving in Australia.

Other Macroeconomic Factors

As well as growth, a number of other macroeconomic factors may also affect saving behavior. The effect of interest rates on saving is theoretically ambiguous as there are potentially offsetting income and substitution effects. Empirical evidence is also mixed: Masson and others (1995) find a positive relationship between real interest rates and private saving, but other studies find no significant relationship. Real interest rates in Australia increased following financial deregulation in the early 1980s, and this could be expected to have boosted saving. The effect of inflation on saving is also unclear, although in time-series regressions inflation is usually positively correlated with saving, as it is probably proxying for measurement biases in national account measures of saving caused by the inflation component in nominal interest payments. Inflation peaked in the mid-1970s, although it remained close to double digits for most of the 1980s, before falling sharply in the early 1990s. While this decline might have acted to boost saving by reducing uncertainty, it is more likely that it will have reduced measured saving as explained above. Transitory movements in the terms of trade should also influence saving behavior as consumption is smoothed in the face of income shocks. However, when the movement is permanent, consumption should adjust to the lower anticipated income level. Masson and others found a positive (Laursen-Metzler type) impact of an increase in the terms of trade on private saving, suggesting that consumption only partially adjusts to changes in the terms of trade. Since peaking in the mid-1970s, the terms of trade fell sharply in the second half of the 1970s, and has since fluctuated around this lower level.

The Tax System

The tax system may influence saving both by changing lifetime wealth and by affecting the rate of return on saving (Boadway and Wildasin, 1994). Under an income tax system, saving is effectively taxed twice—once on the income from which the saving is made, and once on the return (if it is included in the definition of taxable income)—whereas a consumption tax applies only once, when income is used for consumption. For a given total tax level, a shift from an income to a consumption tax therefore reduces the wedge between the pre- and post-tax rate of return on assets. There are two more reasons why income taxes may be detrimental to aggregate household saving. First, income taxes are generally progressive, and they therefore have a greater impact on high-income households—generally the high savers.¹² Second, the working-age population—which comprises the high-saving age groups—pay the bulk of direct taxes. Indirect taxes, in contrast, are more evenly distributed across income and age groups. Callen and Thimann (1997) find the direct tax burden (as measured by the ratio of direct taxes to total tax revenue) to be negatively correlated with household saving, while the indirect tax burden does not have a significant impact on saving. As such, for any given level of revenue, a shift from direct to indirect taxes should have a positive impact on saving. Australia relies heavily on direct taxes, with more than one-half of general government revenue being raised through this source. The United States, Canada, and New Zealand have tax structures similar to Australia. The position is significantly different in many European and Scandinavian countries, where less than one-third of tax revenue is raised from direct taxes (Figure 4.6, top panel). The heavy reliance on direct taxes in Australia may have contributed to the low saving rate (Pender and Ross, 1994; Access Economics Pty Ltd, 1996).¹³ The tax structure has changed little in recent years, however, suggesting that this cannot help explain the decline in saving in Australia, although it may help explain the lower saving level.

View Full Size

Structure of the Tax and Transfer Systems

Sources: IMF, Government Finance Statistics; and OECD, Analytical Database.¹Net transfers, defined as gross transfers net of social security contributions paid by households, indicate the amount of transfers financed from general tax revenues.

• Download Figure
• Download figure as PowerPoint slide

The Coverage, Generosity, and Financing of the Social Security System

The impact of the social security and welfare system on individual saving behavior is likely to depend on a number of features of the system. The greater the coverage of benefits—for example, for pension, unemployment, and health expenditures—the lower the need for private provision in these areas. More generous benefits will lower the need for private provision. Generosity in this context has several dimensions: the value of the benefit payments (defined by the replacement ratio, i.e., the entitlement as a percent of previous earnings); the length of time over which payments are available; the certainty with which people regard the future payment of such benefits; and the general availability of such payments; that is, the extent of “means- and activity-testing.” Finally, the financing of the social security and welfare system may also affect household saving. The system can be financed either from general tax revenue, in which case the benefits are paid out of the general budget, or from specific social security contributions, in which case benefits are generally paid through a social security system that is separate from the budget, but that may receive some budget transfers if contributions fall short of expenditures. The means of financing—taxes or contributions—may be important for household saving because tax financing shifts the financing burden toward higher-income earners relative to contribution financing. Social security contributions are generally levied as a fixed proportion of income, whereas taxes are generally progressive. Contributions also generally begin at very low income levels and are capped at high incomes, whereas income taxes generally exempt very low incomes and are not capped at high incomes. Since low-income households have been found to save little (in many industrial countries the saving rate of households in the lowest income quartile is virtually zero or even negative), and saving rates rise strongly with incomes, a higher reliance on tax financing may lower aggregate household saving.

Callen and Thimann (1997) look at the role of gross and net transfers from the government in influencing household saving. Gross transfers are defined as all government benefits paid directly to households through the budget or the social security system. Net transfers are defined as gross transfers net of social security contributions paid by households, and measure benefits financed from general tax revenue rather than from contributions. In a system that relies heavily on contribution finance, such as those in continental Europe, gross transfers may be high, but net transfers may actually be quite low (Figure 4.6, middle and bottom panels). Callen and Thimann’s results indicate that while gross transfers appear to be an important determinant of household saving when there is a time-series dimension in the data (as in panel data estimation), net transfers are significant in explaining variations in household saving across countries. A possible interpretation of this difference is that over a period-to-period horizon households respond to changes in gross transfers as they affect their current incomes, whereas, over the longer term, they realize that they have to finance the additional payments through higher contributions and only react to the net amount.

In Australia, the welfare system is relatively well targeted and does not appear generous by OECD standards. Still, welfare expenditures have risen sharply since the mid-1970s. Evidence also indicates that while cash transfers are well targeted at lower-income groups, expenditures in areas such as health and education are less well focused on those groups. As far as the financing of the social security expenditure is concerned, there are no specific social security contributions and all expenditure is financed from general revenues. The expansion of the welfare system and the nature of its financing are likely to be important factors behind the low and declining level of household saving in Australia. Other features of the Australian social security and welfare system may influence saving incentives, but they are not easily captured in an aggregate international analysis. This is particularly true in terms of the impact of means-testing, the effectiveness of work tests, and the duration over which some benefits are available.

The Demographic Structure

Based on the life-cycle hypothesis, individuals save more during their working life than in retirement. Therefore, the higher the proportion of people in retirement age, the lower is expected saving. Panel data studies on industrial countries tend to find a significant impact of demographic variables on saving (see Bosworth, 1993; and Masson and others, 1995). In Australia, there have been some noticeable shifts in the demographic structure over the past three decades. As a ratio to the total working-age (15-64) population, the number of people over the age of 65 has increased, while the number of people in the 45-64 age group—likely to be the high savers—declined during the 1970s, but has increased over the past decade (Figure 4.7). For Australia, Morling and Subbaraman (1995) find a significant positive effect on household saving from the proportion of the working-age population in the 45–64 age group (the expected high-saving cohort in the life-cycle model). A related issue is the impact on household saving of the increasing trend toward early retirement in Australia.

View Full Size

Factors Affecting Household Saving

(Year ending June)

Source: Australian Bureau of Statistics; and Reserve Bank of Australia.

• Download Figure
• Download figure as PowerPoint slide

Financial Deregulation

In many countries, liberalization of consumer access to credit has eased liquidity constraints faced by households. As these liquidity constraints previously limited the ability of households to borrow against future expected income and thus raised their need to accumulate assets to finance larger expenditures, their easing could provide a temporary stimulus to consumption and lead to lower household saving (see Bayoumi, 1993). In Australia, financial deregulation has led to a large increase in the outstanding stock of household debt, particularly since the early 1990s (Figure 4.7, bottom panel).

Policies to Raise Private Saving

To raise private saving, as well as to ensure the adequate funding of retirement and to reduce future budgetary pressures from the aging of the population, the government is phasing in a system of compulsory pension (“superannuation”) contributions that will coexist with the public pension and private voluntary schemes.¹⁴ The system will be privately operated and fully funded, and replacement rates will depend on the rate of return earned on contributions. The government has also announced the introduction of incentives for private saving through the tax system. Given that Australia’s population is younger than in many other industrial countries, it has moved relatively early to start to address the problems associated with the aging of the population.

Traditionally, the system of retirement income support in Australia has consisted of a publicly provided age pension and tax-advantaged voluntary private savings. The primary aim of the public pension is poverty alleviation. As such, the pension provides a flat-rate income for retirees, rather than being linked to previous contribution records, and is subject to both an asset and an income test to determine eligibility. The government is legislating a commitment to no less than 25 percent of average male weekly earnings for single retirees and 40 percent for couples. The public pension is paid directly from the government budget, rather than from social security contributions. In the mid-1980s, about 85 percent of those who met the age requirement received income from the public pension, but with the greater use of income and asset tests this has now declined to 81 percent, with around two-thirds getting the full pension. At around 3¼ percent of GDP, gross pension expenditure for the aged in Australia is much lower than in most other industrial countries (about 10 percent of GDP in Denmark, the Netherlands, and the United Kingdom, 12 percent of GDP in France and Germany, and about 15 percent of GDP in Austria and Italy) because the pension system is less generous and, as it is means-tested, less comprehensively available.

Voluntary superannuation has also been an important source of retirement income for many higher income earners and public sector employees. This voluntary system covered around 40 percent of employees in the mid-1980s. Voluntary contributions are subject to concessionary tax treatment, although the extent of these concessions was reduced during the 1980s.

Efforts to increase private provision for retirement began in the mid-1980s. The Accord agreement in 1986 introduced provisions into industrial awards that required employers to contribute 3 percent of employee earnings into a superannuation fund. If such funds were not already in existence, “industry” funds were set up that were jointly controlled by employer and union representatives. These industry funds are generally managed by private fund management companies. Award-based superannuation raised total private coverage to around 80 percent of employees by 1992. Because the coverage of the award system is less than universal, and contributions of 3 percent of award wages were not sufficient to generate a substantial increase in retirement incomes, the government announced a significant expansion of compulsory superannuation in the 1991/92 budget with the introduction of the Superannuation Guarantee Charge. The scheme required employers to contribute at the rate of 4 percent of employee earnings from July 1992 (3 percent if the employer’s payroll was less than $A 1 million). The contribution rate increased from 6 to 7 percent in July 1998, and is scheduled to rise to 9 percent by 2002/03.

In the 1997/98 budget, the government announced some important changes to the existing superannuation system, including: greater choice for employees as to the superannuation fund their money will be invested in; the opting out from contributions for very low-income workers who, with their employer’s agreement, can choose to receive wages or salary in lieu of employer contributions; and incentives to defer claiming the public pension for up to five years after eligibility is reached.

The government has also announced the introduction of a broadly based savings rebate available through the tax system. It will be available in respect of superannuation contributions and/or income up to an annual cap of $A 3,000.

The Outlook for Private Saving

The impact of the superannuation system on private saving will depend on a number of factors. First, much depends on the degree to which voluntary household saving is reduced as a result of the increase in compulsory retirement saving. The impact on household saving will be determined mainly by the number of liquidity-constrained households that are unable to consume more of their current income. It further depends on how important saving motives other than retirement provision have been (e.g., consumption smoothing, saving for education). Various studies have attempted to quantify the offset in voluntary household saving and have found it to be in the range of 30–75 percent (see Fitzgerald and Harper, 1993; Covick and Higgs, 1995; and Morling and Subbaraman, 1995). The Australian Treasury estimates discussed below assume different offset rates for different income groups, ranging from 5 percent for the lowest decile to 50 percent for the highest decile.

The effect on aggregate private saving is further determined by where the incidence of the employer contributions falls. If it falls on employers through higher employment costs (the impact on costs is somewhat mitigated because contributions are tax deductible) then it will likely either reduce corporate saving, reduce employment, with a consequent negative effect on household saving, or some combination of the two. If it falls on employees through lower wage growth, then household saving will be lower than it otherwise would have been.

A further factor concerns the preservation of funds in the system. Benefits in private superannuation schemes only have to be “preserved” in the fund until age 55, and the means-tested public age pension is not available until 61 for women (although this is being raised to 65 by 2014) and 65 for men. Hence, there is an incentive for individuals who have accumulated moderate amounts of superannuation savings (i.e., enough to reduce entitlement to the public pension, but not enough to substantially exceed the public pension) to retire early, run down these savings, and then qualify for the public pension when eligible. Alternatively, given that the primary residence is excluded from the asset test for the age pension, withdrawn superannuation funds may be invested in owner-occupied housing. These may both represent a source of leakage from the system, although the scale of the effects is unclear. To reduce this potential leakage, the preservation age is to be raised to 60 by 2025, and some measures are being taken to increase the attractiveness of annuity benefits relative to lump-sums.

Given the number of uncertainties involved, it is extremely difficult to estimate what impact the superannuation scheme will have on private saving. Australian Treasury projections estimate that the impact will be significant. Private saving is projected to increase by around 2 percent of GDP by 2001/02 and by 4 percent of GDP by 2019/20, while national saving is expected to increase by 1¾ percent and 3½ percent of GDP over the respective periods (see Gallagher, 1997).

International evidence on the effect of compulsory pension schemes on private and national saving is inconclusive. Some early studies (Kopits and Gotur, 1980; and Datta and Shome, 1981) report that changes in compulsory saving are fully reflected in changes in aggregate private saving. Fry (1992), however, found that in Malaysia contractual savings do not induce households to save more, but, instead, are offset by lower discretionary saving, and that they reduce corporate saving and in net terms reduce aggregate private saving. For Singapore, Husain (1995) finds that increased saving in the compulsory pension fund is largely offset by reductions in voluntary saving, although he suggests that prior to the liberalization of withdrawal criteria the impact of compulsion on total saving may have been greater. Faruqee and Husain (1995) largely confirm this finding for Singapore, but find little connection between saving in the pension fund and overall private saving in the case of Malaysia.

The impact of the new tax incentives on private saving behavior is also uncertain. The available empirical evidence generally suggests that tax incentives on one particular type of saving simply encourage a reallocation of saving toward the tax-favored instrument, without resulting in any increase in overall private saving. Furthermore, because the tax incentives themselves lead to a loss of government revenue, national saving may actually decline. Some recent evidence from the United States (see, for example, Hubbard and Skinner, 1996), however, has suggested that there is some positive effect on private saving from these tax incentives. Australian Treasury projections assume a positive effect on private saving, and a small positive effect on national saving (see Gallagher, 1997).

Other structural reforms, including measures to raise potential growth and to increase efficiency and competition in the financial sector, should also help to boost private saving over the long term. They are discussed in the following chapters.