Last modified: 23 October 1997 Maintained by: John QuigginJohn Quiggin
Quiggin, J. (1992), 'Fightback and One Nation: A comparative analysis' Australian Tax Forum 9(2), 12754.
I would like to thank Bruce Chapman, Julian Disney, Rick Kuhn and Nancy Wallace for helpful comments and criticism.
Not since the 'Battle of the Plans' in the 1930s has the debate over economic policy been so central to the political choices facing Australians. The Fightback package is easily the most elaborate specification of an alternative economic policy offered by an Opposition party in Australia. The government's response is an alternative package, titled One Nation. Although One Nation is in many ways less substantial and significant than Fightback, the scope of the competing programs is, at first sight, impressive.
Both programs set out tax scales to apply over the period from now until 1996. Both provide details of expenditure over a similar period. Finally, both are supported by projections of the main macroeconomic variables over the next five years (and, in the case of Fightback, as far as 2000). The parallel with the Battle of the Plans is apt in other ways. The competing plans have been brought forward against a background of mass unemployment, high foreign debt and severe balance of payments problems.
Despite, or perhaps because of, the level of detail in the competing plans, the general reaction has been one of bewilderment. As would be expected in political statements, the packages themselves are presented so as to display the benefits as prominently as possible, while hiding the costs. For example, both packages make use of comparisons between the present tax scales and those promised in 1996. Such comparisons are meaningless unless they take account of 'bracket creep', the phenomenon by which inflation pushes wage earners up the tax scale, raising both average and marginal tax rates.
The confusion has been compounded by the release, simultaneously with One Nation, of Treasury and Finance analyses of the Fightback package. These critiques were prepared in late 1991, but their release was delayed by the ALP leadership change. As a result, they are based on macroeconomic projections that are quite different from those used for the One Nation package. In addition, they contain internal inconsistencies.
The object of is paper is to provide a comparison of the packages on the basis of consistent assumptions regarding wages and prices, employment and gross domestic product (GDP) growth. By 'consistent', I mean that any differences in the behavior of economic variables should be accounted for by differences in the packages themselves. For example, the Fightback package predicts a once-off 4.7 per cent increase in the consumer price index (CPI) as the result of the introduction of the goods and services tax (GST). This increase would be included in the analysis of the Fightback package. However, there would be no corresponding increase in prices in the analysis of the One Nation package. However, the same underlying rate of inflation will be used for both analyses.
Any projection is open to criticism and unlikely to borne out in reality. However, in an exercise of this kind, the critical need is for consistency. It does not matter much whether the rate of inflation is 3 or 5 per cent, provided the same rate is used throughout the analysis.
Both packages are compared to a 'base projection.' In this base projection, the settings of government policy are, as far as possible, left unchanged in real terms. For example, it is assumed that pensions and benefits are unchanged in real terms and that the income tax scales are indexed for inflation, so that a taxpayer with constant real income pays the same real income tax (and has the same after-tax real income) in 1996 as in 1991.
1. The Base Projection
The base projection employed here is derived primarily from the One Nation package, but is broadly consistent with Scenario A in the Fightback package. This scenario involves a sustained economic recovery from 1992-93 onwards with economic growth rates of around 4 per cent. It is assumed that the division of national income between wages and profits continues to shift towards profits so that total wages rise by 3 per cent a year in real terms. Real wages are assumed constant and employment is assumed to grow by 3 per cent per year.
Inflation is assumed to be a constant 3.5 per cent from 1992-93 onwards. All quantities in the scenarios presented here are stated in real terms. However the rate of inflation is important for two reasons. First, the analysis of the two packages involves estimates of the revenue gains arising from bracket creep. Second, the real burden of interest on outstanding public debt rises with unanticipated declines in the rate of inflation.
Government expenditure is assumed to be unchanged in real terms, and therefore to decline steadily as a proportion of GDP (and also on a per capita basis). Savings arising from the gradual reduction in unemployment will be offset by population growth. Tax rates are assumed to be adjusted for inflation, but otherwise to remain unchanged.
Unemployment in this scenario remains high. With employment growth of 3 per cent a year, unemployment can be expected to fall by about 0.5 per cent a year in the recovery phase, because much of the employment growth will be absorbed by rising labor force participation. If employment growth is sustained at 3 per cent, the decline in unemployment may accelerate slightly to 1 per cent a year toward 1995-96. If rates of output and employment growth are lower than those projected here, as many analysts expect, unemployment rates may be expected to remain even higher.
The budget deficit is estimated at $9 billion (or 2.5 per cent of GDP) in 1991-92. As tax revenue rises with increasing economic activity, real revenue rises. Since expenditure is held constant, the deficit is expected to decline gradually. The deficit is, in many ways, the critical parameter in the analysis, not because of its inherent economic significance, but because both parties are committed to the view that continued budget deficits must be avoided. Following this logic, both have presented projections of the implications of their policies showing a return to budget balance by 1995-96. Conversely, to the extent that the projections made here indicate a continued budget deficit under each of the policy packages, it is likely that the packages will be modified to incorporate further expenditure cuts and/or tax increases.
No projection is made of the current account deficit (CAD). According to the currently dominant view, the possibility of an increase in imports and hence the CAD, represents the main constraint on potential recovery. This view has been contested by a number of economists, who argue that the CAD should not be a target of government policy. There is insufficient space to cover this issue here.
Base projections for the key economic variables are presented in Table 1.Table 1- Base Projections
a: Per cent
b: $billion, 1991-92
c: Per cent of GDP
The Fightback Package
The Fightback package is by far the larger and more ambitious of the two. Its stated program involves the abolition of a number of existing indirect taxes (wholesale sales taxes, payroll taxes, import tariffs, petroleum and other excises (except for excise on alcohol and tobacco) and the training guarantee levy) and their replacement by a single goods and services tax (GST) levied at a rate of 15 per cent on most items of personal expenditure. In addition, the package proposes large cuts in personal income tax funded partly by the excess of GST revenue over the revenue from existing indirect taxes, and partly by proposed net expenditure cuts of $4 billion per year (1 per cent of GDP) and partly by bracket creep, of which more later. Finally, large scale privatisation of government enterprises is proposed.
If the calculations in the package are accepted, the tax and expenditure cuts balance each other. Thus, the deficit will follow approximately the same path towards balance as in Table 1, but the final tax and expenditure shares of GDP will be about 1 per cent lower.
A much rosier picture appears if the revenue from privatisations is included as current income in the budget balance. In this case, there are large surpluses commencing with the 1993-94 budget. As is noted in the Fightback package, Mr. Keating used this accounting device to bolster the surpluses of the late 1980s. However, he was correctly criticised at the time for this practice, and it is no more legitimate when adopted by the Liberal and National parties.
An assessment of the Fightback package raises three main issues. First, what is the change in the distribution of the tax burden. Second, where does the burden of expenditure cuts fall. Third, do the tax and expenditure cuts add up, as stated, to produce a zero net impact on the budget.
The price of goods currently subject to the 30 per cent WST will change very little as a result of the GST. On the other hand, goods and services that are currently exempt can be expected to rise in price. Even in these cases the price would not usually be expected to rise by the full 15 per cent. There are two reasons for this. First, very few goods escape indirect tax entirely. For example, although food is exempt, the building and fittings of the supermarket in which it is sold are not. The price we pay for food incorporates the overhead costs of the supermarket and hence an element of tax. The second reason is that some of the benefits from the proposed abolition of payroll tax are expected to flow on to consumers in the form of lower prices. The full 15 per cent rise might be expected to flow on to personal services. These have previously been exempt, involve limited overheads and are typically provided by small firms that are below the payroll tax threshold.
It is estimated in the Fightback package that the changes will result in an overall increase of 4.4 per cent in the CPI. The impact on individual households will depend on two factors. The first is the proportion of household income that is currently consumed. A household that currently consumes its entire income will pay more in GST than one that consumes only 75 per cent of its income. Low income households usually spend a higher proportion of their income than high income households, so the tax will be regressive.
The second factor is the items on which income is spent. Households that spend a large proportion of their income on food, clothing and services will pay more in additional tax than those spending income on items that were already heavily taxed, or that remain exempt under the GST. The most important example in the first group is food which makes up a much larger proportion of the expenditure of low-income households. The decision to include food in the tax base at the full 15 per cent rate, contrary to the practice of most other countries with a GST, increases the regressivity of the tax.
This may be seen by dividing the nation's households into income deciles. For 1984, the ABS Household Expenditure Survey (HES) shows that households in the bottom two deciles allocated 21 per cent of expenditure to food (excluding meals out and take-away). Households in the top two deciles allocated only 13 per cent. In addition, expenditure for the bottom deciles exceeds 120 per cent of income, whereas, for the top two deciles, expenditure is about 70 per cent of income. Hence, expenditure on food, expressed as a proportion of income, is almost three times as high for the poor as for the rich. It follows that a tax on food will bear three times as heavily on the poor as on the rich.
On the other side of the equation, the biggest tax reduction is for luxury motor vehicles and items that remain exempt include private school fees, private health care expenditure and overseas holidays.
A variety of methods may be used to calculate the impact on different households. Most of these rely primarily on the HES. The Fightback package incorporates estimates from the STATAX package, which uses HES data to compute the impact of taxes on a wide range of different household types. Unfortunately, although the package includes detailed estimates of the burden of the existing WST by household type (Tables 5.2 and 5.4), no comparable tables are given for the GST.
For working individuals and households, the main offset for the consumption tax consists of cuts in income taxes, which are to be phased in two stages. In the first stage, from 1 October 1994, the threshold for taxation will be raised from $5400 to $7000. The 20 per cent marginal rate currently applying to income from $7000 to $20 000 will be reduced to 16.2 per cent, and the rates of 38 and 46 per cent applying to income from $20 000 to $50 000 will be reduced to 30 per cent. This tends to favor low and middle income earners. However, the second stage of cuts, from 1 January 1996, the marginal rate on income from $50 000 to $75 000 is cut to 36 per cent and the top marginal rate is cut from 47 per cent to 42 per cent.
A key question is whether the income tax cuts announced in the package are sufficient to offset the increase in indirect taxes. In order to assess this, it is necessary to take account of bracket creep. This is the fact, that, under a progressive scale, as incomes increase with inflation, the average and marginal tax rates rise. To keep the effective tax rates constant it is necessary to index them. The alternative, preferred by all governments is to announce periodic tax cuts that actually do no more than return the proceeds of bracket creep.
This desire is evident in the Fightback package. Whereas all other estimates are presented in real terms (a highly desirable practise), the tax scales are presented in nominal terms. If real terms were used, tax rates on incomes below $50 000 would increase after October 1994, because of the absence of any compensation for bracket creep.
Table 2 shows the tax cuts in Fightback adjusted for bracket creep. Columns 1-3 are as published in Fightback. Column 4 contains estimates of nominal income after 4 years inflation at an annual rate 3.5 per cent. Because of the effects compound interest this implies a cumulative price rise of 14.7 %. Column 5 shows the tax payable under the Fightback (1995-96) schedules. In Column 6, this amount is deflated back to 1991-92 dollars. The real tax cut, shown in Column 7, is the difference between this and the tax paid as in Column 2. In Column 8, the tax cut is expressed as a percentage of 1991-92 income. The smallest cut, 1.7 per cent, goes to individuals earning $20 000 per year. The largest cut, 8.0 per cent, goes to those earning $60 000 per year.
|Income Level||Tax currently payable||Tax payable under Fightback||Apparent tax cut||Income adjusted for inflation||Tax payable under Fightback||Tax payable under Fightback||Real tax cut|
|(Col 2 - Col 7)||% of income||Real tax cut net of GST||Real tax cut net of spending cuts||Real tax cut net of spending cuts and GST|
|$91-92||$91-92||$95-96||Inconsistenta||$95-96||$95-96||$91-92||$91-92||% of income||% of income||% of income||% of income|
The final column is an attempt to match this cut against the increased tax burden associated with the GST. The estimated impacts are derived from the work of Professor John Freebairn of Monash University. As can be seen from the Table, the net impact is negative for all individuals earning below $35 000 per year and positive for those earning more than $35 000. A more detailed assessment could be provided using a model such as STATAX.
This basically regressive impact of the GST is partially offset by two other features of the Fightback package. The first is the proposed cap on income tax deductions for superannuation. The second is the 100% increase in Family Allowance which is received in full only by households with incomes below $30 000 per year and at a 50 per cent rate by households with incomes between $30 000 and $40 000 per year (It is not clear if these figures are indexed for inflation). On the other hand, the Family Allowance Supplement, which is paid to the very poorest households (with income below $19 300) will be increased by only 4.4 per cent, and it is not clear whether the threshold will be indexed. Overall, the redistribution of the tax burden among taxpayers without children will be highly regressive. The redistribution of the tax burden among taxpayers with children will be generally regressive, but only moderately so. Three will be a slight shift in the tax burden from households without children to households with children.
The long-run impact of a (partial or complete) switch from income tax to a GST is to increase the tax burden on the young and old and reduce the tax burden on the middle-aged. In the idealised models normally used to analyse tax policy, this redistribution has no impact on welfare at any stage in the life cycle. Young households simply increase their borrowing to maintain their optimal level of consumption. In middle-age, households increase their saving to allow for the need to pay higher levels of GST in old age. These models are based on the false assumption that borrowing and lending are costless or, equivalently, that the bank margin between borrowing and lending is zero.
A more realistic model, with bank margins of 5 per cent, yields quite different results. First, the young are indeed worse off, since they are deterred from borrowing by high real interest rates. Second, the distortion generated in lifetime consumption patterns means that lifetime welfare is reduced, even though the lifetime tax burden is unchanged.
In general, the biggest holders of financial assets are the retired and those approaching retirement. The Fightback package includes provisions for compensation for this wealth effect. However, they are confined to persons over 60. Hence, those approaching retirement will lose significantly and receive no compensation.
In combination with another Fightback initiative, the tax-free savings scheme, the wealth levy effect creates incentives for financial manipulation. The tax-free savings scheme involves an exemption from income tax for new deposits with financial institutions. Hence, people who now have such deposits have a double incentive to convert them into real assets or shares, thereby avoiding the wealth levy, then to convert them beck into financial deposits, thereby qualifying for the tax-free savings scheme.
On these assumptions, business gains $5.8 billion from the abolition of the payroll tax of which $2.9 billion is passed on to consumers. A further $1.1 billion is clawed back in company tax on the increased net profit, leaving a gain of $1.8 billion. However, the increase in the company tax rate from 39 to 42 per cent brings in $1 billion, so the net reduction in business taxes is only $800 million.
Even this is an overestimate. The package includes an estimated $1.2 billion in revenue from taxation of foreign tourists and the 'black economy.' On the standard assumption that Australia is a 'small country' in the world hospitality market, the taxes on foreign tourists will be borne entirely by the Australian tourist industry. The 'black economy' refers to businesses that will be brought into the tax net by the need to claim GST rebates on their inputs. The estimate of $1.2 billion is exaggerated. But even if the correct figure is only half this, the net reduction in business taxes falls to a trivial $200 million.
The effect of the Fightback package is, in general, to redistribute the tax burden from big business to small business. The gain of $1.8 billion accrues entirely to firms above the payroll tax threshold ($500 000 in most states). However the increase in company tax is paid by all companies big and small.
These effects are reinforced if we drop the assumption that WST, excise and GST are passed on in full. The WST and excise fall predominantly on the products of large companies. It seems reasonable to assume that the profits of these companies will be increased by the replacement of a 20 per cent (or higher) WST and excise (with no offset for inputs) by a 15 per cent GST (with full offset for inputs). Conversely, the lightly taxed services sector is dominated by small firms. These firms will lose as a result of the switch from WST/excise to GST. Similarly the money extracted from the tourist industry and the black economy will come primarily from small firms.
In summary, the 'Fightback package' will have almost no effect on the total tax paid by the business sector, but will shift the tax burden from large to small firms. It is of interest to ask why the representatives of small business have generally supported the package. One important reason is that the income tax cuts in the package strongly favor the upper income group (above $35 000 per year). Most self-employed and small business people do not fall into this group (see Fightback Table 10.3, p. 153). However, both the more successful businessmen who tend to be (self ?) appointed representatives of small business and the professionals (economists, tax accountants, PR advisors) they hire have high incomes and are clear beneficiaries of the personal income tax changes.
The Fightback package is unusual in that, although it gives a wide range of exemptions (inclduing financial services, gambling, private education and health and overseas holidays), the traditional categories of food and clothing are taxed at the full rate. This is not particularly serious in the case of clothing. Although clothing is often represented as a basic need, it is not really subject to satiation (consider the expenditure of Imelda Marcos on footwear!). Clothing makes up much the same proportion of expenditure for all income groups. However, as has already been shown, the taxation of food is highly regressive.
Furthermore, the standard efficiency arguments for uniformity are very weak in the case of food. Definitional problems appear to be minimal and the aggregate elasticity of demand for food is very low, indicating that any efficiency loss associated with a differentially low tax rate would be minimal.
The general approach of the Fightback package has been to take a very generous approach with politically influential groups, such as the financial, gambling and private health and education industries. These groups have argued that, on second-best or practicality grounds, their products should be exempted or zero rated. This generosity has meant that unless the rate of tax was to be set above the 15 per cent seen as politically feasible, food had to be included in the tax base.
A particularly important deviation from efficiency in the design of the tax is worthy of mention. An ideal consumption tax would tax Australian expenditure on overseas tourism (an import) but would exempt foreign tourist's purchases in Australia (exports). In practice, this is impossible, and all GST systems adopt the reverse implying a distortionary tax on one of the most employment-intensive of export industries. Many GST systems offer schemes under which at least part of the expenditure of foreign tourists is exempted. The Fightback package has no elements of this kind, presumably for revenue reasons. On the other hand, even domestic travel incidental to overseas tourism is exempted from the tax. As well as being highly distorting, this element of the package will exacerbate balance of payments difficulties.
A striking featur
Despite the rhetoric about the exceessive expansion of the state, the Fightback package shies away from any explicit commitment to a reduction in public services. Rather it is assumed that major savings can be made by eliminating the unholy trinity - fraud, inefficiency and waste.
An example is given by the Defence portfolio. The Fightback document proposes gross savings of $500 million from cuts to defence administration (of which $300 million are to be directed toward increases in the combat element of the forces) but does not specify a single area that is to be cut. This is illustrated by Table 3. The $500 million in cuts has been allocated to the category of Fraud, Inefficiency and Waste. The $300 million in promised new defence capabilities has been allocated to the category of Changes in Programs
Social Security &Welfare:
Privatisation (Public Debt Interest)
Cut in State grants
Fraud, Inefficiency & Waste
Changes in Programs
a Cuts are consequent on abolition of fuel excise and reflect various offsetting expenditures (eg exemptions for farm use) that will no longer be necessary. New expenditure is compensation for wealth effects of the GST.
Leaving aside the areas of housing health and social security, the cuts in Federal spending that can be attributed to changes in programs (cessation of functions or funding cuts for specific programs) total only $865 million, compared to over $1.5 billion to be saved from the elimination of fraud, inefficiency and waste. The cuts in specific programs are more than offset by new spending commitments. Even if the increase in Family Allowances (costing about $1 billion) is treated as a tax measure, there is no net reduction in spending arising from changes in programs.
The extent to which fraud, inefficiency and waste can be cut is doubtful, especially bearing in mind that the base scenario incorporates a continuation of the tight restraint on expenditure that has characterised the Hawke and Keating governments. It seems far more likely that the required cuts will be achieved by (as yet unspecified) cuts in functions.
The largest single area of cuts is the social security portfolio where $2.4 billion in cuts are sought to offset $2.6 billion required to compensate the recipients of pensions and benefits for the price impact of the GST. Here again, the emphasis is on the elimination of fraud and inefficiency. Hardly anyone who is now in receipt of a government benefit will become ineligible, although nearly everyone will face more complex and strenuous tests for eligibility.
Unemployment benefits provide the most important area of proposed cuts. The Fightback package proposes a complex set of work tests for the first nine months of unemployment followed by a requirement to undertake 'work for the dole' or private sector employment at 'training wages'. (At the same time, substantial savings in administrative expenditures are expected). But no benefit is actually abolished. Further, the package assumptions that up to 130 000 people will go off unemployment benefit as a result of these measures but that no currently employed workers will be displaced. This is despite the fact that the projections of total employment growth, presented as part of the package do not include anyhting like 130 000 extra jobs. The savings claimed from the proposed measures are overestimated, probably by about 50 per cent as indicated in the Department of Finance critique of the package.
The cuts of $400 million in housing expenditure are to be implemented by forcing the states to sell off public housing. This should not be counted as an expenditure saving but as an asset sale.
The estimated proceeds from privatisation are to be used to pay off debt, reducing the government's interest bill by $1.3 billion. But the fact that privatisation will mean the loss to the government of the profits of Telecom, the Commonwealth Bank and other government business enterprises (GBEs) is simply ignored. The Fightback document points out that some of this will be recouped as company tax. For the rest, it states blandly (p325) "to the extent that there is a net loss to the budget ... the three year program outlined in this chapter will enable any shortfall between decreased dividend receipts and public debt interest to be adequately covered." The Dept. of Finance review of Fightback suggested a loss to revenue of $330 million. However, in keeping with standard Federal government accounting practices, this estimate takes account only of dividends paid by GBEs. The correct measure of the loss to the public sector as a whole includes profits that are retained and reinvested. An estimate of this total amount cannot be obtained from the Budget papers but it appears that net of company tax the profits of GBEs were of the order of $1 billion. Because of the failure to take the loss of GBE profits into account, the net saving from privatisation will be much smaller than estimated in Fightback. If the proceeds from privatisation are lower than estimated, there may be a net loss. Similarly, the saving of $400 million in the housing area is to be obtained by forcing the states to sell off public housing. This should not be counted as an expenditure saving but as an asset sale.
The absence of any substantial cuts in programs makes it almost impossible to assess the impact on households of the expenditure cuts proposed in Fightback. However, a best attempt is as follows. First, we may disregard the savings on public debt interest from privatisation and the tax compensation item. The savings in public housing can be assumed to fall ultimately on the occupants of public housing and the social security savings on benefit recipients. Of the program changes about $1 billion in new spending is associated with the family allowance changes. These can be allocated to households with incomes below $40 000 that have children. The remaining program changes approximately cancel out. Hence it may be assumed that they will have no net impact on the welfare of the average household.
It remains to assess the impact of the proposed cuts not associated with the abolition of, or reductions in funding to, any particular program. This remainder consists of the cuts in state grants, the savings allocated to fraud, inefficiency and waste and the reduction in health spending (total $3831 million), offset by the $180 million spent on the new first home owners scheme for a net amount of $3651 million. In the absence of any real information on these cuts, it will be assumed that they fall evenly on the population as a whole. The average cut is about $215 per capita or about $410 per taxpayer.
A net benefit for taxpayers may be obtained by subtracting $410 from the real tax cut given in column 7 in Table 2. The results are significant primarily at the bottom end of the scale. They suggest that the combined effect of the income tax and expenditure changes will leave most low-income taxpayers (those below $20 000 per year) no better off, even before the impact of the GST is taken into account. As regards recipients of social security, the aged will generally be better off and the unemployed and occupants of public housing worse off.
There are also significant problems on the revenue side. These relate to estimates of the likely revenue from a GST. The estimates in the Fightback package are derived as follows. An ideal consumption tax would fall on private final consumption expenditure measured at $230 billion in 1990-91 by ABS. However, the proposed GST exempts or zero rates certain items (overseas travel, rents, building construction, private health and education spending and items such as alcohol, tobacco and gambling that are taxed separately). On the other hand, the expenditure of foreign tourists in Australia, which would be exempted by an ideal consumption tax, is included in the tax base. Taking these deductions into account, the base for the GST is about $169 billion. Overseas experience suggests that in practice about 80-95 per cent per cent of the theoretically available tax will in fact be collected, suggesting a revenue of $20-$25 billion. The Fightback package assumes that tax collection will amount, not to 100 per cent of the theoretically available tax, but to 105 per cent or $27.2 billion. The extra revenue is expected to come from the 'black economy'.
The estimate that the black economy is between 5 and 10 per cent of GNP is probably reasonable. The problem is with the claim that the GST will bring the black economy into the tax net, by forcing people to register for the first time. Not only is this supposed to raise around $1 billion in extra GST revenue, but it is assumed (p. 670) to provide an additional clawback in income tax. The empirical evidence from other countries does not support this assumption. Officials from Canada and New Zealand have denied that their GST systems had produced big gains from the black economy.
The reasoning behind the Fightback package displays a misunderstanding of what the black economy is. It does not, for the most part, involve people working completely outside the tax system. Such people were never very important and, since the Tax Summit in 1985, most of them have been brought into the tax net by the Tax File Number system. The real black economy involves people like plumbers and shopkeepers who report all their expenses to the tax authorities (as deductions from taxable income) but take some of their income in the form of unreported cash sales. After the GST is introduced, they will keep on doing the same thing.
The Department of Finance estimates that revenue from the GST will be about $2.5 billion less than that claimed in the Fightback package. This is based on a fairly optimistic collection figure of 95 per cent, and is, in my view a lower bound to the error in Fightback. On the expenditure side, the on-budget savings have been overestimated by about $1.2 billion, representing the over-estimates of social security cutsd and the failure to take account of the loss of dividends from GBEs. Thus, the estimated deficit arising from the Fightback measures is around $3.7 billion.
However, as has been noted above, the budget balance is misleading. In addition to the dividends from GBEs the package will entail the loss of retained earnings (about $600 million). Further, about $450 million of claimed cuts in the housing and education portfolios in fact represents either the sale of assets or the bringing forward of income. Hence the true deterioration in net public sector savings is of the order of $5 billion.
With all its faults, the Fightba
The response to the income tax cuts in Fightback is straightforward. It reflects a political judgement that middle income earners are the group most likely to change their votes in response to income tax cuts and that marginal, rather than average, rates are the main focus of concern.
In One Nation, the most attractive elements of the Fightback tax cuts are duplicated, while more expensive elements are ignored. The marginal rate of tax for income from $20 000 to $36 000 is cut to 30 per cent, matching the marginal rate offered by the Fightback package. For income from $36 000 to $50 000, the marginal rate is cut from 46 to 40 per cent. There are no cuts on taxes applying to incomes below $20 000 or to marginal rates above $50 000.
This may be seen as a repetition of Labor's successful 'marginal seats' strategy in the 1990 elections. Upper income earners are ignored since they are mostly committed Liberal voters. Lower income earners are ignored because they have 'nowhere else to go'. As was shown above, this group will be hit hard by the consumption tax and the expenditure cuts proposed in the Fightback package. The middle income 'swinging' voters are given cuts in marginal tax rates matching those offered by Fightback.
As was observed above, bracket creep plays a central role in the financing of Fightback. This is even more true of One Nation. Since there are no expenditure cuts or new taxes, the tax cuts in One Nation are funded out of a combination of bracket creep and deficit financing. The role of bracket creep is illustrated in Table 4.
Columns 1-4 are as published in One Nation. Column 5 has estimated nominal income after 4 years inflation at 3.5 per cent (a compounded increase of 14.7 %). Column 6 is the tax payable under the One Nation (1995-96) schedules. In Column 7 this is deflated back to 1991-92 dollars. The real tax cut is the difference between this and the tax paid as in Column 2. Column 9 gives the tax cut as a % of 1991-92 income.
There is an increase in the real tax burden for all income earners with current taxable incomes less than $25 000 per year, and a real tax cut for everyone above this figure. Thus, the key group, with incomes between $25 000 and $35 000 a year, gains from the One Nation package, but would be net losers under the Fightback package. The value of the tax cuts is about equal to that in the Fightback package for incomes between $40 000 and $45 000. Above this level, the cuts in Fightback are greater than in One Nation.
The standard focus on marginal rates can be misleading, a fact that is carefully exploited in One Nation. This can be illustrated by the fact that, although those with incomes above $50 000 receive no cut in their marginal rate, they are better off under the One Nation package. This is because they pay less tax on the portion of their income between $20 000 and $50 000. Another aspect of marginal rate illusion can be seen with incomes from $20 000 to $25 000. Income earners in this range receive a real cut in income tax under Fightback, although this is more than offset by the GST and expenditure cuts. Even though One Nation offers the same marginal tax rate, its tax cut is insufficient to offset bracket creep. This is because there is no cut in tax on the first $20 000 of income.
The One Nation package also includes a limited response to the increase in family allowance proposed in the Fughtback package. This consists of a once-off cash payment to families in recent of Family Allowance, fruitlessly timed to coincide with the Wills by-election campaign, and a $3 (10 per cent) increase in the Family Allowance supplement going to low income families.
The other pressure affecting One Nation is the need to make some response to the recession. The Fightback package is designed for introduction after a change of government in 1993 and, quite properly, includes no discussion of short-term fiscal policy.
In assessing this aspect of One Nation, it is useful to examine the general stance of government policy during the recession. In all previous recessions, at least since World War II, the government has attempted to stimulate the economy through a combination of tax cuts and expenditure increases, and has also offered various forms of direct relief to the unemployed. In the present recession, the government, backed by a consensus of élite opinion, has resisted any measures of this kind. Indeed, 'pump-priming', formerly a Keynesian metaphor to suggest the beneficial nature of fiscal stimulus in generating a self-sustaining recovery, has become a dirty word. Advocates of stimulus have been at pains to deny that they have any form of 'pump-priming' in mind.
The only remaining influence of Keynesian orthodoxy is that the government, and élite opinion generally, has been fairly relaxed about the expansion in the deficit caused by the recession through declining tax revenues and increasing unemployment benefits. These factors used to be referred to in Keynesian parlance as 'automatic stabilisers', since they ensured a desirable fiscal stimulus in response to an economic downturn. The budget balance has gone from a surplus of $4 billion before the recession to an estimated deficit of $10 billion for 1991-92, a decline of $14 billion, almost entirely as a result of the recession.
The period leading up to the One Nation package is not the first time the government has faced public pressure to abandon its 'hands-off' stance. The March Industry statement of 1991 was also a political response to the recession, which was then expected to end late in 1991. Broadly speaking, the strategy of the Industry statement was to give the minimum possible stimulus to the economy consistent with the political imperative to 'do something'. The industry statement included about $133 million of expenditure directed at unemployment relief, equal to about 1 per cent of the impact of the automatic stabilisers. Although some aspects of the Industry statement were controversial, the decision to give only a derisory response to the unemployment problem received almost unanimous support from the élite.
In One Nation, the expenditure component is a little larger. The Family Allowance Supplement increase and a range of smaller social security and education measures lead to a permanent increase in spending of $0.5 billion. In addition, the package includes three elements that might be regarded as counter-cyclical and are presented with a certain amount of Keynesian rhetoric.
The first is a program of infrastructure spending, totalling $770 million in 1992-93 and $306 million in 1993-94. The second is a package of unemployment relief measures, similar in scope to those in the Industry statement ($150 million in 1992-93 declining to $45 million in 1993-94 and to zero in 1994-85, when the unemployment rate is anticipated to be around 9.5 per cent). The third is a set of industry assistance and tax relief measures of which the most notable are the reintroduction of accelerated depreciation allowances (one of the key elements of the Fraser government's plans for an 'investment-led recovery') and the provision of tax relief to the banks in return for an understanding that they would expand lending to finance investment.
The combined impact of all of these measures is a net fiscal stimulus of $1.6 billion in 1992-93 and $1.3 billion in 1993-94, or about 10 per cent of the automatic stabilisers referred to above. Indeed, the impact of the One Nation measures is dwarfed by the revision of the estimated budget deficit that has taken place since One Nation was announced. When One Nation was released at the end of February 1992, the estimated deficit for 1991-92 was $7.3 billion. In May, it was revised to $9.6 billion, a change of $2.3 billion.
On the assumptions presented here, the One Nation package will have a long-run negative budgetary impact of about $6 billion. Since the baseline projection has the budget returning to balance in 1995-96, the deficit under One Nation would be about $6 billion or 1.5 per cent of GDP in that year. By contrast, the projections presented in the One Nation package suggest that budget balance would be achieved in 1995-96. This is primarily because the package employs an optimistic base projection. In the absence of the One Nation package, a large surplus is projected for 1995-96.
In summary, the fiscal outcome is precisely the opposite of what would have been supported by Keynesian orthodoxy. Over the next year or so, there is almost no fiscal stimulus. Marginal increases in expenditure are offset by the decision to give no compensation for bracket creep over this period. On the other hand, the large tax cuts offered for 1994 and 1996 when the economy will presumably be near a cyclical peak, imply the generation of a structural budget deficit.
It may reasonably be argued that the Keynesian model is largely irrelevant. In particular, the notion of a cyclical peak involving unemployment levels of around 8 per cent seems almost self-contradictory when viewed from a Keynesian perspective. However, given the derisory nature of the labor market programs proposed in the One Nation package, it is only its moderately Keynesian rhetoric that gives the appearance of any concern about unemployment.
The Fightback package does not directly address the issue of unemployment. It is claimed, however, that the associated labor market reforms will reduce unemployment (Ch. 19). This claim is backed up by scenarios derived from the Murphy model. However, the latter are essentially the result of imposing the desired result on the model through parameter changes. The overseas experience of radical reforms of the kind proposed here suggests the likelihood of a short-run jump in unemployment combined with a permanently higher base level of unemployment.
Some issues covered in the Fightback package are not addressed in One Nation. Most importantly, whereas Fightback sets out an explicit schedule for privatisation, One Nation does not discuss this issue. It seems reasonable to assume that, if re-elected, Labor will undertake further privatisation, but at a slower rate than that envisaged in Fightback. As argued above, if the device of including asset sales in the budget balance is disallowed, the net impact of privatisation on the budget will be minimal.
In many ways, the similarities
One difference is, of course, the GST, and the associated abolition of a range of indirect taxes. This proposal is much more radical than the incremental expansion and rationalization of the WST system that has taken place under Labor.
More generally, there is a difference in tone between the two statements. The Fightback package is characterised by a wholehearted commitment to the set of ideas generally referred to as 'economic rationalism'. By contrast, One Nation represents an attempt to distance the Labor government from these ideas, even though they continue to dominate most aspects of policy.
Because of these differences, it seems reasonable to infer that the response to deficits of the kind projected here will differ depending on which party wins the next election. Once elected, I expect that a Hewson government would find the extra $5 billion in expenditure cuts necessary to finance its tax cuts, even if this meant abandoning commitments made in Fightback. By contrast, a Keating government would be more likely to accept a somewhat larger budget deficit or to delay the implementation of the promised tax cuts.
Overall, though, the most important conclusion yielded by a study of these packages is that neither will have more than a marginal impact on the economic problems facing Australia and particularly on unemployment. Both parties have essentially accepted the conclusion that unemployment rates of above 8 per cent are here to stay and that nothing can, or should, be done about it.