Tran-Nam, Binh --- "Assessing the Revenue and Simplication Impacts of the Governments Tax Reform" [1999] JlATax 24; (1999) 2(5) Journal of Australian Taxation 329

	Home \| Databases \| WorldLII \| Search \| Feedback Journal of Australian Taxation

ASSESSING THE REVENUE AND SIMPLIFICATION IMPACTS OF THE GOVERNMENT S TAX REFORM

The electoral victory of the Federal Coalition Government in October 1998 on the platform of a GST-based tax reform was unprecedented in Australian federal politics. The GST Bills, introduced by the Government on 2 December 1998 and passed by the Senate in July 1999, have sharply divided the opinion of tax academics and tax practitioners. There appears to be equally strong support for and opposition to the GST amongst Australian tax experts. This article presents an assessment of the Federal Governments tax plan on two criteria of good tax policy, namely revenue adequacy and simplicity. Its main conclusion is that the GST-based tax reform will increase the operating costs of the Australian tax system by about $1 billion annually. Implications of the Governments tax plan to Commonwealth-State financial relations will also be discussed.

1. INTRODUCTION

The re-election of the Howard Government in October 1998 on the platform of a substantial, new tax was unprecedented in Australian federal history. The electoral victory of the Federal Coalition Government signifies that the Australian public resistance to a Goods and Services Tax ("GST") has been successfully overcome. It also ensures that Australia will enter the new millennium with a substantial indirect tax reform after three failed attempts in 1978, 1985 and 1993.^[1] For its successful handling of the tax reform process, the Coalition Government, and Mr Howard in particular, should be commended.^[2] It remains to be seen, however, whether the GST-based tax reform will deliver the many economic benefits that the Federal Government has been promising.

The six GST Bills, introduced by the Government on 2 December 1998, are massive in terms of sheer size and contain hundreds of pages of legislation.^[3] These Bills were amended and passed by the Senate in July 1999. The principal changes, demanded by the Australian Democrats, included the exclusion of basic food from the GST. The Senate inquiry had again highlighted the fact that the proposed GST has sharply divided the opinion of tax academics and professionals. There appears to be equally strong support for and opposition to the GST amongst leading tax experts.

The principal aim of this article is to present a preliminary assessment of the Government's tax plan, including the new Commonwealth-State financial arrangements arising from the tax reform. The assessment is based on two criteria of good tax policy: revenue adequacy and simplicity, with a particular emphasis on the simplification impact of the tax plan. This is not because simplicity is more important than other principal criteria of good tax policy, namely economic efficiency and equity. But it is because (i) tax reform debates in Australia and overseas have focused on the issues of distribution of tax burdens and economic prosperity at the expense of simplicity, and (ii) the effects of the tax reform on the operating costs of the Australian tax system can be quantified with a reasonable degree of accuracy. Since the GST in Australia will commence on 1 July 2000, the analysis is necessarily based on a combination of plausible theoretical speculations, relevant international and domestic empirical evidence, and simulation of Australian data using general equilibrium modelling.

The remainder of this article is organised as follows. Part 2 examines the revenue adequacy aspect of the Government's tax plan. It is argued that the Government's estimates of the GST revenue seem to be somewhat optimistic. The simplification impact of the Government's tax reform is explored in Part 3. Despite the Government's claim to the contrary, the tax plan is estimated to increase the recurrent operating costs of the Australian tax system by about $1 billion annually. Thus, complexity (in the operating costs sense) is indeed a major weakness of the new tax system. Part 4 discusses a number of issues relating to commonwealth-state financial relations under tax reform. Some concluding remarks are then given in the final section.

2. REVENUE ADEQUACY OF THE TAX REFORM

The rapid adoption of the GST (or value added tax ("VAT") as it is known in Europe and Asia) around the world in the last 30 years has been due principally to the requirement of trade zone groupings, the advocacy of international agencies and its powerful revenue-yielding capability. Clearly the revenue-raising capacity of the GST is fundamental for the success of the Australian Government's tax plan. To understand how the GST revenue is calculated, consider a hypothetical example based on Australian macro data in 1996/97.^[4] Assuming that the level and the distribution of consumption remain unchanged under the original GST, Australian private consumption expenditure (inclusive of GST but exclusive of the Wholesale Sales Tax ("WST") and State taxes to be abolished) in 1996/97 would be $317 billion. This consists of about $200 billion taxable, $74 billion input taxed and $43 billion GST-free. Adding to the taxable base ($199.3 billion), the taxed portion of inputs for residential and financial services ($48.8 billion) gives a grand total GST base (post-GST) of $238.1 billion. The GST revenue would then be:

GST Revenue = GST Rate x Grand Total GST
Base (post - GST)
_________________________
1 + GST Rate
= 0.10 x 248.1
___________
1.10
= $22.6 billion

The GST revenue ($23 billion) and other positive indirect effects of tax reform ($2 billion)^[5] would replace the WST ($13 billion), business stamp duties ($2 billion), FID/Debits tax ($2 billion), offset cuts in excise duties (alcohol, cigarettes, gambling and fuel - $3 billion) and income tax ($5 billion). If the Government plans to give income tax cuts worth $13 billion, then the introduction of the original GST would result in a net negative effect on the Government budget of about $7 billion.

It has been estimated by the Federal Treasury that a 10% GST with minimal exemptions and zero-rating (including the exclusion of basic food from the GST) will generate $24.97 billion, $28.16 billion and $28.71 billion in 2000/01, 2001/02 and 2002/03, respectively.^[6] As indicated above, these estimated revenues are sufficient to replace the WST and a range of State indirect taxes, and to permit some reductions in excise duties and income tax. The Government's estimates are based on a number of assumptions including an annual GDP growth rate of 3.5%, the positive effects of the GST on economic activities and employment, no significant changes in consumers' spending pattern, and a 95% compliance rate of the GST base. GST revenue collection is very sensitive to these assumptions, and the limited theoretical and empirical evidence available suggests that the Government's forecasts may be somewhat optimistic. Leaving aside the assumed GDP growth rate, which is essentially non-GST driven, it is worthwhile to examine other assumptions more closely.

2.1 GST and Small Businesses

Given that economic inputs are not perfectly divisible, it is inevitable that the burdens of complying with the GST fall disproportionately on small businesses (in the sense that tax compliance costs as a percentage of turnover declines steadily as turnover increases). This is particularly true in the case of a transaction-based GST where large businesses are far better equipped to deal with the GST-induced record keeping requirements than small businesses. As a result of this actual or perceived high compliance costs, small businesses may reduce sales in order to stay out of the GST system. This hypothesis has been confirmed by a study by the Business School of the Open University in the UK. This study finds that nearly 18% of companies surveyed in Britain are deliberately avoiding sales to stay below the ₤50,000 threshold for VAT (the corresponding GST threshold for Australian businesses is $50,000). Thus, the paperwork burdens of the GST on small businesses may, without appropriate policy responses, act as a barrier to economic growth and employment.

2.2 Household Production and GST Avoidance

Unlike goods, many service items can be produced through household production for own consumption, or exchanged through bartering. Thus, the introduction of a broad-based consumption tax, whether a multi-stage GST or a single-stage Retail Sales Tax ("RST"), will encourage a shift of these services from the formal (that is, taxable) to the informal (that is, non-taxable) sector. This is particularly true for those services where the main component of costs is the labour of the suppliers, eg, haircuts, childcare, educational coaching, bookkeeping and so on. Such changes in consumers' behavior in order to avoid GST could give rise to a substantial shrinkage in the GST revenue base associated with the taxation of services.

2.3 Cash Economy and GST Evasion

It is often argued by the Government and proponents of tax reform that a GST is self-policing and it will flush out the cash economy. This holds true, in principle, to all suppliers except for those in the last stage of the distribution process (eg, restaurant owners or tradespersons). If these businesses operate on a cash-economy basis (in order to evade income tax) before the introduction of a GST, they will still have the opportunities to do so after the GST. In fact, these businesses will now have an additional incentive not to charge GST to their customers in order to avoid fallen demands.

A recent study shows that the cash economy in New Zealand has grown from 7% of GDP in 1970 to 11.3% in 1994, an increase of over 60%.^[7] While it is not possible to attribute this growth to New Zealand's tax reform in the mid 1980s, it is evident that the imposition of a GST and the corresponding reduction in income has done nothing to reduce the cash economy in New Zealand. The empirical evidence from Canada is more clear-cut. Various studies have consistently indicated that there has been a significant growth in the cash economy and tax evasion following the introduction of a GST in 1991.^[8]

In summary, there are warranted reservations about the Government's argument that the GST and a range of other measures will reduce the growth of the cash economy in Australia. Two further revenue-related issues deserve mention.

2.4 GST and the Aging of Population

It has been argued that broadening of the consumption tax base provides an effective mechanism for coping with the aging of the population that has been taking place in the Australian and most developed economies. In the words of Lester Thurow:

In fact, a central part of the Federal Treasury's arguments in favour of changing the tax mix is that the present array of indirect taxes will not raise sufficient revenue to pay for Australia's future public-sector requirements. However, simulation of Australian data by Dixon and Rimmer using the Monash model has found no evidence to support the Treasury's contention. In their base-case forecasts, current indirect taxes grow at about the same rate as GDP, even in the absence of any increase in the indirect tax rates.^[10]

3. SIMPLIFICATION IMPACT OF THE TAX REFORM

In its pre-election tax reform package, the Federal Government recognised the complexity of the Australian tax system and devoted a whole chapter to simplification of the tax system. Apart from the replacement of the WST by a GST, there were a number of other simplification measures. The Government then claimed that its tax plan will "deliver higher economic growth and more jobs for Australia as a result of...lower tax compliance costs...".^[11] The principal aim of this part is to examine validity or otherwise of the Government's claim.

3.1 Tax Operating Costs: Some Conceptual Issues

Tax simplicity (complexity) is defined here as the ease (difficulty) with which the correct tax liability can be determined, collected and enforced. The ascertainment of tax liability involves not only taxpayers but also tax advisers, tax administrators, tax lawmakers and judges. This concept of effective simplicity can be measured in terms of the effort (that is, value of the resources) expended by the society in raising some amount of tax revenue. It is thus important to develop a sound theoretical framework for estimating the operating costs of a tax system.

3.1.1 Components of Tax Operating Costs

The operation of any modern tax system typically involves five broad types of activities:

In Australia, the tax policy divisions of the Federal and State Treasuries are normally responsible for the design and planning of tax policy, including the preparation of a new tax or the amendment/abolition of an existing tax. In addition, the academia, business community, and welfare and religious organisations may also try to influence the governments' tax policy through academic writings, submissions, proposals and lobbying activities.

At the federal level, the Office of Parliamentary Counsel is responsible for the drafting of necessary tax legislation with the assistance from the ATO. Both Houses of the Commonwealth Parliament spend much time and resources in debating, amending and turning tax proposals into statute law or legislation. Similar but simpler activities are also performed at the state level.

The ATO administers the Federal tax system by interpreting the law (via determinations and rulings), collecting taxes, assessing and auditing tax returns, enforcing compliance, recovering tax debts, refunding tax overpayments, forecasting tax revenue and resolving tax objections.^[12] At the state level, tax revenues are collected by State tax offices, usually a division within the State Treasury.

Not only taxpayers (individuals and firms) but also non-profit institutions and the budget sector of the federal, state and local governments have to expend resources in meeting the requirements laid upon them by the tax system. Tax compliance is broadly defined to include both computational and tax planning activities.

There are various mechanisms for resolving tax disputes in Australia. Apart from the ATO (before the dispute is taken further) and the Administrative

Appeals Tribunal ("AAT"), the Federal Court and ultimately the High Court have jurisdiction to finalise substantive tax disputes. Although State courts do not have jurisdiction to hear substantive tax disputes, they have jurisdiction in tax debt recovery disputes.

The conventional practice at present defines tax operating costs as the sum of administrative and compliance costs. While it is true that these two components constitute the bulk of the operating costs of the tax system, it is conceptually unsound to ignore the costs associated with the legal framework of the tax system. It seems worthwhile attempting to quantify the costs associated with tax policy design and planning, tax drafting and enactment, and tax dispute resolution. As the parliaments, government departments and the courts perform a variety of functions, this will be a difficult task of cost allocation.

The need to consider the tax operating costs as a whole has more than pedagogic relevance. It is well known that the various components of the operating costs of a tax system are transferable. For example, if the drafting of sophisticated tax legislation is underfunded, this may lead to more costly tax collection or tax disputes later. This reinforces the need to measure the operating costs of a tax system in totality rather than by its individual components.

3.1.2 Net and Social Tax Operating Costs

It seems sensible that in discussing the impact of tax simplification, the costs of operating the tax system must be:

First, the operation of the tax system generates not only costs but also benefits to the society as a whole. Some of the most frequently cited operating benefits are:

Thus, the relevant tax operating costs must refer to the difference between the gross operating costs and operating benefits of the tax system. In practice, these offsetting benefits are very difficult to quantify in money terms and typically are omitted in empirical studies.

Secondly, in the case of tax compliance activities, it is possible to distinguish between the public sector and the private sector components of tax compliance costs. This distinction arises because there are compliance benefits to taxpayers that vanish at the societal level. These include:

The cash flow benefits enjoyed by taxpayers may be thought of as interest-free loans to taxpayers and thus viewed as costs to tax authorities. This means that cash flow benefits to taxpayers represent a zero-sum transfer from the public to the private sector, which do not reduce the compliance costs to the economy, assuming taxpayers and tax authorities have the same interest rate.^[13] Similarly, assuming that the resources employed in tax compliance activities are perfectly mobile, tax deductibility benefits to taxpayers cancel out revenue losses to tax authority. Since it is costly to administer a subsidy program, the public sector costs of government subsidy will typically exceed the benefits received by the private sector.

The above discussion suggests that it is not useful to view compliance costs as private sector costs as conventionally done in the literature.^[14] It would be more rigorous to disaggregate the social tax compliance costs as follows:

The above framework implies that tax planning or tax minimisation is wasteful from a societal point of view. As rational maximising agents, taxpayers engage in tax planning because the costs of tax planning are presumably lower than corresponding reductions in their tax liabilities. But the reductions in their tax liabilities will be matched by losses of the same magnitude in government tax revenue, that is, tax planning represents a zero-sum transfer from the public to the private sector. At the same time, the social costs of tax compliance will increase by the same amount as the costs of tax planning (assuming that if taxpayers do not engage in tax planning, professional tax advisers can still find employment elsewhere). By the virtue of the same argument, tax enforcement is also socially wasteful.

3.1.3 Measuring Tax Simplicity (Complexity)

The tax simplicity criterion of good tax policy, first expounded systematically by Adam Smith,^[15] seeks to minimise the operating costs incurred in raising a given level of tax revenue. Translating into symbols, in a static (that is, one-period) model of the economy, the tax simplicity criterion strives to minimise the unit-free O/T ratio where O and T are the tax operating costs and tax revenue, respectively. Tax simplification is then said to take place when this ratio falls as a result of a tax change. It is worthwhile to note two conceptual problems associated with this kind of approach.

First, dividing O by T adjusts for such factors as size of the economy (other things being equal, the larger the number of taxpayers, the higher are the operating costs and so are tax revenues) and business cycle (when the economy is booming, both tax operating costs and tax revenues tend to rise). But it fails to take into account the effect of overall tax severity. As a mental exercise, imagine a small rise in the rate of a GST. It is plausible to assume that the operating costs do not grow as fast as tax revenue. In this case, it seems awkward to call the decline in the O/T ratio "tax simplification".

The second, and more complex problem, is that the tax system operates not in a single year but over many years or decades. Any tax change involves both commencement (once-off) and recurrent (on-going) costs, and it is uncertain whether or not the ratio of the recurrent operating costs over total output will remain constant over time. Under the simplifying (and plausible) assumption that the ratio O_r/Y (where O_r stands for the recurrent operating costs) does not fluctuate

too widely over time, one can, in principle, ignore the commencement (fixed) costs and compare the two ratios arising from the two tax alternatives under consideration.

If the O_r/Y ratio fluctuates widely over the years, it is far more problematic to compare the simplicity of two tax alternatives. In the simple but unrealistic case where both tax systems have the same lifetime, it is possible to define the aggregate tax operating costs and aggregate tax revenue over the lifetime of a tax by making use of the present value concept in cost-benefit analysis. Even so, the data requirements are simply prohibitive. If two tax systems have different lifetimes, comparing their relative simplicity is practically impossible.

3.2 The Governments Tax Plan and its Own Simplification Impact Assessment

3.2.1 Simplification Measures in the Tax Reform Package

The key feature of the Government's tax plan with significant implications on tax operating costs is the replacement of the WST and a range of State taxes by a 10% GST with minimal exempted and zero-rated categories. The Tax Law Improvement Project will be abolished and the Income Tax Assessment Act 1997 will itself disappear ultimately in the integrated tax code.^[16] Other important proposals regarding business taxation include:

There are also some discussions regarding personal income taxation (eg, abolition of work-related expenses and non-lodgment of tax returns) but these are unlikely to be implemented, at least in the near future.

3.2.2 The Government s Own Impact Assessment

In its Regulation Impact Statement ("RIS") accompanying the main GST Bill,^[18] the Federal Treasury has provided a comprehensive statement on the impact of the Government's tax reform on the administrative and compliance costs of the Australian tax system. For its openness, the Government must be congratulated. However, a careful analysis reveals that the Federal Treasury has not only underestimated the magnitude of total compliance costs but also employed an invalid concept of tax compliance costs.

In the RIS for the GST, the ATO estimates that the costs of administering the GST will be (in 1999/00 prices) $340 million, $351 million and $301 million for 1999/00, 2000/01 and 2001/02, respectively.^[19] Since basic food is now excluded from the GST, overseas evidence suggests that the new administrative cost estimates will be higher than those indicated above. Note that there will also be some reductions in tax administration costs resulting from the abolition of WST and a range of State taxes. The ATO's estimates reflect its plan to employ an extra 3,000 additional staff, mostly graduates, to cope with the GST. The estimates for 1999/00 and 2000/01 appear to have included start-up costs so that, despite a steady growth in the consumption tax base, the total administrative costs of the GST are projected to fall in the third year of operation.

The Government's analysis of the recurrent compliance costs of the GST in the year 2001/02 is set out in page 8 of its RIS. There are many serious problems with the Government's own assessment.

(i) Government's estimate of gross recurrent costs of GST (WST) is too low (high)

Based on a NZ study by Sandford and Hasseldine, the Government estimates that the recurrent gross compliance costs of the GST are $1,195 per firm or $1,910 million for 1.6 million registered firms in 2001/02. Sandford and Hasseldine found that the average annual gross compliance costs of GST in NZ was NZ$1,290 (= $452,957,000/351,236) in 1990/91.^[20] Since the GST was first introduced in NZ in 1986, it seems reasonable to take this estimate as representing recurrent compliance costs only. Assuming conservatively that compliance costs grow by 1.7% per annum, this would give rise to an average cost of NZ$1,552 in 2001/02. Using average weekly earnings ("AWE") data for New Zealand and Australia, it is found that for the period 1991-1998, NZ$1 labour cost in New Zealand translates to AUS$0.8842 in Australia.^[21] Based on this relative wage information, an estimate of recurrrent GST compliance costs of $1,372 per firm is obtained for Australia in 2001/02.

The Government also estimates the net (that is, the private sector) compliance costs of the WST at $830 million. This seems to be a mistake in interpreting the results of an ATO-sponsored study of the compliance costs of the Federal tax system by a team at the Australian Taxation Studies Program ("ATAX"). The ATAX study found the gross compliance costs of the WST to be $737 million in 1994/95.^[22] Using an annual cost inflation rate of 1.7%, the estimate of the gross WST compliance costs would be $830 million in 2001/02. The net WST compliance costs, after taking tax deductibility and cash flow benefits into account, would be substantially lower than this figure.

(ii) The conceptual framework is invalid from a public finance point of view

The Government's analysis is based on the business sector's point of view and is not concerned with the resource costs to the entire economy. It is thus invalid from a public finance standpoint. For example, the Government can make the recurrent GST compliance costs to the private sector as small as it wishes by providing businesses with an on-going subsidy of appropriate magnitude. This will reduce the GST compliance costs to businesses, but not to the economy.

Despite using an invalid conceptual framework and underestimating (overestimating) the recurrent gross compliance costs of the GST (WST), the Government still arrives at a positive increase in compliance costs to be borne by businesses. This contradicts a previous statement by the Government in the same document that its tax reform will "... reduce compliance costs for business".^[23]

3.3 Assessing the Simplification Impact of the Governments Tax Plan

3.3.1 General Assumptions

In order to assess the simplification impact of the Government's tax reform, it is necessary to make a number of general assumptions:

A1The ratio O_r/Y remains stable over time for both the current and new tax systems.

Under this assumption, it is possible to examine how the ratio O_r/T changes as a result of the tax reform.

Many economists believe the economy-wide output effect of the GST will be negligible or even negative. One reason is that, since the formal sector is likely to be more efficient than the informal sector, the substitution of some services into informal activity may lead to resource wastage. Piggott and Whalley have constructed a model of the Canadian economy which indicates that the base broadening from the move to a GST in Canada in 1991 was in fact efficiency worsening.^[24] In the Australian context, estimates of the efficiency gains of the GST range from -3% of total employment in the short-run^[25] to 1.8% of GDP in the long-run.^[26] If the GST-induced growth rate in GDP is close to zero, it is possible to focus on the change in the recurrent operating costs of the Australian tax system as an indicator of simplification.

A3The reductions in personal income taxation and excise duties have trivial effects on the operating costs of the tax system.

Under this plausible assumption, it is possible to compare the GST versus the WST and some State taxes that the GST will replace.

3.3.2 Commencement Costs

The Australian society has devoted a substantial amount of time and energy in discussing, debating, opposing and promoting the GST, particularly in the lead up to the Federal elections in 1993 and 1998. This involves many sectors of the Australian economy, including the Government and opposition parties (both at federal and state levels), business, welfare and religious organisations, labour unions, tax academics and practitioners, and the media. In particular, the Federal Treasury has expended considerable resources to study the impact of the GST on different sectors of the economy under various scenarios.

Following the re-election of the Howard Government, the Office of Parliamentary Counsel aided by The Treasury and the ATO drafted the six massive legislative packages of GST Bills. These Bills were passed but the GST laws are likely to be further amended in the future.

The start-up administrative costs of the GST can be deduced from the ATO's cost estimates reported in the Government's IRS. They are at least equal to $39 million in 1999/00 prices.

The start-up costs of the GST, including the costs of registering the business, learning about GST legislation, printing new stationery, purchasing new equipment such as cash registers or computers, acquisition of tax software, etc, will be reasonably high. To reduce the compliance costs of the GST, the Government plans to promote greater use of electronic connections between the ATO and businesses. According to the ATAX Business Taxpayer Survey, only 27.9% of taxpayers in 1994/95 used computers for tax purposes in their businesses.^[27] Thus, businesses may incur considerable start-up computer costs to take advantage of the Government's initiative.

The correct approach in treating commencement costs involves using the present-value arithmetic to convert start-up costs into equivalent annual costs, eg, if the interest rate is 5% per annum and the start-up costs are $5 billion, then the annual equivalent of the start-up costs is $0.25 billion (= 5% x 5).

3.3.3 Operating Benefits and Other Simplification Measures

The GST will certainly demand more stringent record keeping from taxpayers and this will produce managerial benefits to them. This kind of benefit is difficult to quantify and only a qualitative assessment is possible. Since large firms tend to have stricter record keeping than smaller ones in the pre-GST situation, more rigorous financial recording will generate relatively more managerial information to small firms. It seems plausible to speculate that small businesses value this sort of information more lowly than big businesses so that the aggregate value of GST-induced managerial benefits to taxpayers may not be too significant.

In principle, the massive amount of GST data received by the ATO will help the Australian Bureau of Statistics to improve its measurement of national output, particularly the GDP by the production ("value added") approach. In practice, it is difficult to put a money value to this kind of benefit.

ABN and PAYG are usually considered to generate beneficial simplification impact on the operating costs of the tax system. However, their benefits should not be overestimated or taken for granted. First, several business taxpayers may operate under the umbrella of the same business entity. For example, a medium-sized business entity may consist of one or more registered companies, a family trust, a supernannuation fund and perhaps even a few partnerships. If each of these business taxpayers is given an ABN, the simplification benefits of the ABN will not be too large.

Secondly, GST is to be remitted to the ATO either monthly or quarterly depending on business size, and this is similar to the arrangements under the WST. Remitting taxes to the ATO quarterly under the PAYG system may increase the paperwork burden to a large number of taxpayers who pay their taxes less frequently under the existing system. These include, for example, lump sum provisional taxpayers (who pay their taxes once a year), large companies (twice a year) and small companies (once a year).

3.3.4 Recurrent Operating Costs

Note that the year of comparison is 2001/02 so that all monetary values are expressed in 2001/02 prices.

Although the costs of GST policy design and planning are basically once off, some elements may be of a recurrent nature. The recurrent elements include, for example:

While the legal/political costs of introducing the GST are once off, the costs of amending the GST in the future will be recurrent.

The main sources of information on the administrative costs of the WST are the financial statements contained in the ATO annual reports. Up until 1996/97, the ATO's costs of tax collection are broken down by tax types (income tax, WST and other taxes/charges). For some reason, this practice ceased in 1997/98. According to the ATO's 1996/97 Annual Report, the costs of collecting WST in 1996/97 were $58.32 million.^[28] Assuming a cost inflation rate of 1.7% per annum, this is equivalent to $63.45 million in 2001/02.

It is not easy to ascertain the collection costs of those State taxes to be abolished. Collectively, the State tax offices cost $175 million in 1995/96 (with the NSW Office of State Revenue contributing more than 30% of the total costs).^[29] NSW is one state for which more detailed information is available. Using the data provided in the NSW Treasury's Budget Papers, the costs of collecting the to-be-repealed taxes were about $15 million for NSW in 1997/98.^[30] Assuming that NSW's share of collection costs is only a quarter of all States, the collection costs of those State taxes to be abolished are estimated at $60 million in 1997/98 or $64.19 million in 2001/02.

Adding the two figures together, the administrative costs of the WST and State taxes to be repealed would be about $123 million. Since the recurrent administrative costs of the GST is $301 million, the new tax system involves an annual increase of $173 million in the recurrent administrative costs of the tax system.

Based on the ATAX study mentioned in Part 3.2.2, the recurrent WST gross compliance costs are estimated to be about $737 million in 1994/95 or $802 million in 1999/00. Regarding compliance costs of State taxes, there is a paucity of empirical evidence. A national study by Pope, Fayle and Chen estimated the compliance costs of the payroll tax at 3.6% of total payroll tax revenue for all states.^[31] There are also two studies on some State taxes, one by the Queensland Chamber of Commerce and Industry^[32] and the other by the Chamber of Commerce and Industry of WA.^[33] However, both of these were based on small-scale business surveys with only state-wide focus and with inadequate explanation of the methodologies used. Clearly none of these limited studies is appropriate for the purpose at hand.

In its IRS, the Government estimates the compliance costs of the State taxes to be abolished at about $220 million to businesses.^[34] Assuming that tax deductibility and cash flow benefits are worth half of that total, the recurrent gross compliance costs of the State taxes to be repealed would be about $330 million. This is a very plausible estimate because the estimated compliance costs/revenue ratio for State taxes (to be abolished) is similar to that of the WST.

The recurrent operating costs of WST and the State taxes to be repealed are thus estimated at $1,160 million (= 830 + 330) in 2001/02. Using the estimate obtained in Part 3.2.2, the GST compliance costs are $1,372 per firm or $2,195.2 million (= 1,372 x 1.6 million) for 1.6 million registered businesses in 2001/02. The net increase in the recurrent tax compliance costs is therefore about $1,035 (= 2,195 - 1,160).

The introduction of a GST in Australia will likely generate more tax dispute resolution than the taxes to be abolished, not least because the GST is new and untried but also because it involves a wide range of goods and services, and an immense number of taxpayers. The costs of tax dispute resolution will be of a recurrent nature.

	WST and some State Taxes	Proposed GST	Net Change
Tax Policy Design, Planning, Law Drafting & Enactment	?	?	? but positive
Tax Administration	128	301	173
Tax Compliance	1,160	2,195	1,035
Tax Dispute Resolution	?	?	? but positive
Operating Costs	1,288	2,496	1,208

It is apparent that the Government's tax plan will increase the recurrent operating costs of the Australian tax system by more than $1 billion annnually.

4. TAX REFORM AND INTERGOVERNMENT FINANCIAL RELATIONS

4.1 Commonwealth-State Financial Relations Under the GST Regime

The Commonwealth-State financial relations will be transformed under the Federal Government's tax reform. The new arrangements, as set out in the Federal Government's pre-election package and the post-election agreement between the Commonwealth and the States, have the following main features:

Under the new arrangement, States will abolish a number of taxes and agree not to reintroduce these, or similar taxes, in the future. The proposed timetable for the repeal of these taxes is:

4.2 Impact of the New Intergovernmental Financial Relations on States

The new intergovernmental financial relations, as stated above, are clearly not the first best option from the State Governments' perspective. A number of reasons lead to this suggestion:

State Treasury officials have consistently argued in favour of tax-base sharing,^[36] and the Federal Government's tax plan does not give State Governments their preferred option.

Under the GST-based tax reform, the State Governments' tax base will be substantially reduced. This means that State Governments will be under increasing pressure to tinker with their remaining taxes and that, in the absence of interstate harmonisation agreements, tax competition between states may be expected to rise.

The problems associated with VFI will be exacerbated, as State Governments will become even more dependent on Commonwealth-State revenue sharing under the proposed tax reform. It is also conceivable that the State Governments may collectively be forced to approach the Federal Government to increase the GST rate.

In its pre-election tax reform document, the Coalition Government claimed that the budgetary position of the States will be "enhanced over time by the new tax system, with the States projected to gain about $370 million in 2003/04, $1.25 billion in 2004/05, with commensurately larger gains in subsequent years."^[37] This claim was disputed by the NSW State Government. In a media release on 11 September 1998,^[38] the NSW Treasury asserted that, over four years, the GST package will leave the States $4.4 billion worse off when compared with the existing grant system, or $3.2 billion worse off when compared with the Government's proposed tax plan. NSW would be the most affected with a loss of $2.8 billion loss to the State's revenue between 2000/01 and 2004/05.

As a result of the NSW Treasury's dispute, the Federal Government has agreed to the principle of revenue neutrality as part of the transitional arrangements. In particular, the Federal Government will "guarantee during the initial period following the introduction of the GST (not less than 3 years) that in each of those years the States and Territories will be no worse off financially in aggregate than they would be under the current arrangements. Payments under this aggregate funding guarantee will be made so that the budgetary position of each State and Territories will be no worse off in each year."^[39]

A study by the present author has lent support to the NSW Treasury's claim that NSW would "lose out" under the GST-based tax reform. Applying the tax plan to NSW in 1997/98, it is estimated that the NSW State Government would be directly worse off by:^[40]

This hypothetical loss would correspond to about 6.5% of the NSW Government's total budget receipts in 1997/98. To compensate for the NSW State Government under the revenue neutrality agreement, it seems plausible that the Federal Government will continue to provide NSW with general-purpose grants from income tax revenue. This may still give rise to two problems in the long-term:

▪ continuing disputes between the Federal and NSW Governments about the extent of the net loss of revenue to NSW; and

▪ the Federal Government may take over some services currently provided by the NSW State Government.

On the positive side, it is theoretically plausible that the proposed GST will increase the remaining State taxes on payroll, motor vehicles and land.^[41] This possible increase in tax revenue will somewhat compensate State Governments for their net losses arising from the new intergovernmental financial arrangements.

5. CONCLUSIONS

Nothing in recent times has divided the opinion of Australians as the GST. After many heated discussions and debates as well as two Federal elections, the GST has continued to polarise the community of tax academics and practitioners. Many experts see it is a timely and inevitable reform that will promote Australia's economic prosperity and growth with some minor equity consequences. Equally many others view it as simply a way of changing the tax mix with no genuine benefits to the economy. It seems, at this stage, not even time can bring about a consensual point of view on the GST.

The preliminary calculations in this article suggest that although tax simplification is long overdue, it has again been overlooked in Australia. Overall, the simplification impact of the Federal Government's tax reform package is likely to be negative. The available evidence contradicts the Government's pre-election promise that its GST-based tax system will "collect revenue in a ... simpler...way"^[42] and post-election claim that the abolition of WST and a range of State taxes will "reduce compliance costs for business and simplify tax administration for both the Commonwealth and the States".^[43] In fact, it is most likely that the recurrent tax operating costs will rise significantly with GDP remaining largely unchanged under the new tax regime.

Having established that complexity is indeed a major weakness of the Government's tax reform, it did not make sense for the Australian Democrats to insist on zero rating basic food. Such actions will only serve to further increase the already substantial operating costs of the Government's tax plan. If equity is a major concern, then it should be addressed by some well-targeted compensation programs. Such income transfers will also involve transaction costs, but these costs are likely to be smaller than the increase in operating costs as result of excluding basic food from the GST base. What is least needed in Australia now is a GST with many exempted and zero-rated categories whose number will grow and grow over the years.

Binh Tran-Nam MEc (ANU) PhD (UNSW) is a senior lecturer at ATAX located within the Faculty of Law of the University of NSW. Binh has previously published in academic journals in Australia, UK, US, Germany, Japan and the Netherlands.

^[*] This article is a substantially revised version of a paper presented at the Australian Tax Teachers Conference hosted by the School of Law, University of Canberra and National Centre for Corporate Law and Policy Research, Canberra, 5-7 February 1999. The author wishes to thank Professor Peter Dixon for his valuable comments. Financial assistance from a small ARC grant (Faculty of Law, University of NSW) and from the New South Wales Taxpayers' Association is gratefully acknowledged. Part 4 of this article is based on a research paper commissioned for and funded by the New South Wales Taxpayers' Association.

^[1] Australian Commonwealth Taxation Review Committee, Full Report (1975); National Tax Summit, Record of Proceedings (1985); Liberal Party of Australia, Fightback! (1991, 1992) and the Coalition's loss at the 1993 Federal election.

^[2] After the 1993 debacle, Mr Howard ruled out a GST as part of the Coalition's policy for the Federal election in 1996. How-ever, it has always been his intention to introduce such a tax. As early as 1984 he wrote "I have long believed that the single most important reform which is needed to the Australian taxation system is the broadening of the taxation base towards a greater reliance on general consumption taxes with a corresponding reduction in our current over-reliance on personal taxation as a source of revenue.": See JW Howard, "Taxation Reform" (1984) 1(1) Australian Tax Forum 12.

^[3] The A New Tax System Bills 1998 (Cth). Some legal experts believe that the same rules can be written in far less number of words.

^[4] For details see C Murphy, "The Implications of the Government's Tax Plan-Modelled Using MM303" 2-4, Forum on Model-ling Australian Taxation, Faculty of Business, University of Technology, Sydney, 10 December 1998.

^[6] P Costello, Tax Reform-Not A New Tax-A New Tax System (1998) 30 and J Howard, Costing of Commonwealth and State Mea-sures, Press Release 31/5/99, 1.

^[8] N Brooks, "Lessons For Australia From the Canadian Experience With the GST: Don't Do It!" in B Tran-Nam (ed), Tax Reform and the GST-An International Perspective (1998) 132-133.

^[10] PB Dixon and MT Rimmer, "The Government's Tax Package: An Analysis Based on the Monash Model" (10 December 1998) 16, Forum on Modelling Australian Taxation, Faculty of Business, University of Technology, Sydney.

^[12] Another main function of the ATO is to collect child support maintenance payments and Higher Education Charges Scheme but this aspect of the ATO's operation does not fall within the scope of the present article.

^[13] CT Sandford, M Godwin and P Hardwick, Administrative and Compliance Costs of Taxation (1989) 23 noted that "whilst the total cash flows cancel, the benefits and detriments may not precisely cancel because of the different terms on which public and private sectors can borrow and lend."

^[15] A Smith, An Inquiry in the Nature and Causes of the Wealth of Nations (1776) Book Five, Chapter II.

^[17] Ibid, 133-134. This article is not concerned with wide-ranging business taxation proposals contained in the Ralph Review on Business Taxation.

^[18] Australian Treasury, Regulation Impact Statement for the Introduction of a Goods and Services Tax, A New Tax System Bills 1998 (Cth).

^[20] CT Sandford and J Hasseldine, The Compliance Costs of Business Taxes in New Zealand (1992) 62.

^[21] Key Statistics, various issues, Statistics New Zealand, Wellington and Year Book, Australia, various issues, Australian Bureau of Statistics, Canberra. The Australian data refers to total AWE of all employees (full-time and part-time) while it is not clear whether the New Zealand data refers to total AWE of all employees or full-time employees only.

^[22] C Evans, K Ritchie, B Tran-Nam and M Walpole, A Report into Taxpayer Costs of Compliance (1997) 56 AGPS, Canberra.

^[24] J Piggott and J Whalley, "VAT Base Broadening, Self Supply, and the Informal Sector" (1998) National Bureau of Economic Research Working Paper 6349.

^[26] Murphy, above n 4, 14. His calculations showed a 1.8% increase in GDP but most of it was from an increase in capital, not an increase in efficiency. The Murphy's estimate of the increase in efficiency was $0.6 billion or 0.13% of GDP.

^[27] C Evans, K Ritchie, B Tran-Nam and M Walpole,A Report into the Incremental Costs of Taxpayer Compliance (1996) 92-93.

^[29] O Gabbitas and D Eldridge, "Directions for State Tax Reform" (1998) Productivity Commission Staff Research Paper 44.

^[30] NSW Treasury, Budget Information 1997-98, Budget Paper No. 2 (1998), A-80.

^[31] J Pope, R Fayle and DL Chen, The Compliance Costs of Employment Related Taxation in Australia (1992)

^[32] Queensland Chamber of Commerce and Industry, Regulatory Costs and Other Burdens, A Survey Report (1996).

^[33] Chamber of Commerce and Industry of Western Australia, A Case Study into the Compliance Costs of Taxation (1996).

^[35] Costello, above n 6 and Australian Treasury, Agreement on Principles for the Reform of the Commonwealth-State Financial Relations, Media Release, 13 November 1998. States in this article mean both States and Territories.

^[36] G Izmir, "State Perspective of Tax Reform" in B Tran-Nam (ed), Tax Reform and the GST - An International Perspective (1998) 41 and J Langoulant, "Revenue Sharing versus Tax Base Sharing", State Taxation: Repeal, Reform or Resignation, ATAX, Sydney, 11 June 1998.

^[38] NSW Treasury, Tax Reform: Impact on NSW State Finances (11 September 1998).

^[40] B Tran-Nam, "The Impact of the GST on NSW State Taxes" 12-18, Australian Taxpayers Association National Conference, 10 March 1999.