Sydney Law Review
‘Simplification’ has been the mantra of tax reformers and tax deformers since the late 1950s and has frequently been cited as a rationale for tax changes in the closing years of the twentieth century and the opening years of the twenty-first. The near universal agreement by tax critics that simplicity is an object of tax reform is not mirrored by universal agreement as to what simplicity entails. For some, simplicity means a tax system that is easy and inexpensive to comply with. For others, it means simple and easy to understand language in the tax legislation. And for others still, simplicity means simple in its effect, with a comprehensive law containing a minimal number of distinctions and exceptions so all arrangements or transactions with similar economic effects will receive the same tax treatment.
Measured against almost any test of simplicity, the Australian income tax law fails abysmally. While the costs of tax administration incurred by government are not high by international standards, study after study shows the tax system imposes higher compliance costs on taxpayers than virtually all other income tax systems and taxpayers almost universally believe the law to be complex. The legislation may well be both the largest in the world in terms of sheer volume and among the most difficult to read and comprehend. Almost certainly, more irrational distinctions based on inappropriate criteria exist in the Australian law than in any other nation’s tax legislation with the result that minute changes in the legal form of a transaction can lead to dramatically different tax consequences.
In the mid-1980s, a government review of the Australian taxation system noted that ‘piecemeal improvements have been made to the system over the years but the point has now been reached where fundamental reforms — rather than further running repairs — are called for.’ By the start of the 21st Century, both the size and complexity of Australian income tax law had grown out of control. The legislation had expanded to many thousands of pages and it continues to grow at an exponential rate — plotted on a graph, the growth curve reveals that if the trend continues, within a few years Australia will be the first nation in the world to boast an infinite number of pages in its income tax law!
This paper explores the phenomenon of complexity in the Australian income tax. The next part of the paper considers the causes of complexity, reviewing the possible contributions of the parties most closely connected to the tax system — the judges who interpret the legislation, the tax advisers who work with it, the drafters who write it, the Treasury that designs it, and the legislature that enacts it as law.
The paper concludes that all parties must bear some blame for the complexity but prime responsibility falls on the legislature on two counts. First, the legislature has failed to eliminate the irrational distinctions in the law to which most complexity can be traced and, second, it has added new layers of complexity by using the tax system as a spending tool to distort market and social behaviour.
The commercial transactions to which income tax is applied are among the most complex arrangements possible and it is inevitable that an income tax will acquire some of the characteristics of the transactions to which it applies. But if an absolutely ‘simple’ income tax in all the senses that word is used cannot be realised, simplification or the taming of the complexity of the current system is well within the realm of the achievable. The third part of the paper considers the two most recent proposals for simplification reform, redrafting the tax law in ‘plain English’ and the adoption of a new tax law based on a ‘tax value method’ in preference to conventional notions of gross income and deductions. It then outlines a very brief proposed prospectus for taming complexity. The paper suggests that in the context of current Australian political reality, the preferable model for reducing complexity in tax law is one based on incremental simplification.
Income taxation in Australia pre-dates the nation itself. Several of the separate colonies levied income taxes and these taxes remained in place after federation in 1900 when the colonies-transformed-to-states were granted the constitutional power to share the income tax base with the newly formed Commonwealth government. Commonwealth income tax was adopted in 1915, allegedly as a war time finance measure, but no doubt partly in response to calls by the rural sector for income taxation of profiteering manufacturers. Those calls were loudest in the west, articulated most often by the disgruntled farmers who supported secession from the new nation. The adoption of Commonwealth income taxation conveniently deflated one of the platforms for the separatist movement.
The Commonwealth tax operated concurrently, if not in parallel, with state income taxes. Different tax bases imposed high compliance costs on taxpayers and led to costly administrative inefficiencies. An attempt in 1922 to harmonise taxes with a new Commonwealth Act and state Acts was a failure. A second attempt leading to the adoption of a new Commonwealth Act and harmonised state Acts in 1936 was somewhat more successful, though these were falling out of synchronisation by the advent of the Second World War. War finance needs prompted the Commonwealth government to usurp the income tax field by raising Commonwealth taxes to the point where taxpayers could not pay both the state and Commonwealth taxes, while promising transfer payments to any state that relinquished its state income tax. This move withstood a number of constitutional challenges and since the mid-1940s only the Commonwealth has levied income tax.
For several decades the Act doubled in size every seven years but the pace of change has increased significantly in the past three decades, as has the complexity of the law. The 1936 Act remains in place today, albeit many times its original size, and operates alongside a 1997 Act, with separate laws applying to particular tax issues such as tax administration and the taxation of fringe benefits.
Blame for complexity in the income tax has been attributed to many parties. The alleged miscreants and their possible roles in weaving the web of complexity in the income tax are reviewed below.
Blaming the judges for the complex state of the Australian income tax has long been a favourite pastime of Australian commentators. The practice certainly peaked in the 1970s and early 1980s when it seemed to many (with some justification) that the High Court’s regular endorsement of blatant and artificial avoidance schemes played a central role in the near disintegration of the income tax system. But to this day, judges remain a popular target to blame for the troubles of the tax system.
The judiciary are commonly blamed in particular for three practices: misappropriation of doctrine by using concepts from other categories of law to solve income tax issues, misapplication of precedent by relying almost reverentially on judgments of UK courts, and abdication of judicial responsibility through reliance on principles of strict literalism. There is some merit to all these criticisms.
(i) Fallacy of the Transplanted Category
Australian common law, like that of all English-speaking jurisdictions, is based on the doctrine of precedent. Fairness is thought to derive from consistency, and consistency, in turn, is thought to flow from the application of the most appropriate precedent to the facts at hand. The challenge is to find the most appropriate precedent and it is here that Australian judges have failed, falling into the trap referred to as the ‘fallacy of the transplanted category’. The fallacy of the transplanted category is the application in one area of law — tax law in this case — of doctrines and precedents deriving from another area of law, yielding results that are clearly inappropriate in terms of the policy objectives underlying the recipient body of law.
Misapplying precedents that seek to achieve entirely different policy objectives is a failing of which Australian judges appear to be guilty with remarkable regularity. To decide which gains constitute assessable income for purposes of the income tax law, Australian courts have turned to trust law precedents that distinguish income gains, payable to life beneficiaries, from capital gains, payable to a remainder beneficiary of a trust. These doctrines were developed to decide between competing claims by income and capital beneficiaries to realised gains, not to characterise the gains as appropriate or otherwise to bear a particular tax levy. To decide whether the transferor or transferee has derived shifted income, courts have turned to property law doctrines on the effectiveness of equitable assignments, doctrines that have nothing to do with economic command of resources and ability to pay tax. To distinguish ‘employees’ from independent contractors for tax purposes, courts regularly turn to tort or industrial law precedents, though the policy rational for distinguishing between employees and independent contractors for, say, vicarious liability purposes bears no relationship to that for imposing withholding tax collections on one category of income earners and self-assessment taxpayer lodgement of taxes on another. To determine the meaning of business ‘goodwill’ for the purpose of applying a tax concession, courts draw on a range of precedents including partnership and contract law doctrines used to evaluate the reasonableness of restraint of trade covenants, an issue that enjoys no correspondence with the subsidy of retirement savings of small businesspersons, the apparent objective of the goodwill tax measure. Examples can be found in all areas of tax law.
The impact of transplanted categories on the complexity of Australian law has been enormous. Without doubt, the transplant that has caused the most complexity is the application of trust law indicia to define income for income tax purposes. The courts seized upon the indicia of income gains for trust law purposes — income gains were gains that were periodic, gains that were anticipated and deliberately pursued by the recipient, receipts that were applied to common living expenses, benefits that comprised cash or assets readily convertible to cash, and so forth — and excluded from the definition of income in the income tax law a vast range of common gains such as discounts on debt, premiums on leases, non-convertible benefits, payments for non-competition covenants, gains from cancellation of debt, and gains realised on the disposal of investment assets.
The judicial concept of income — euphemistically labelled ‘income according to ordinary concepts’ by the courts — was not only far too narrow to sustain a viable income tax base but was inherently uncertain, being based on shifting indicia such as taxpayer’s subjective intent with regard to transactions and investments. Over the decades, the indistinct boundaries between the judicial concept of income gains on the one hand and other gains on the other, along with the equally indistincts on the outgoing side, have been at the heart of virtually all major tax avoidance schemes and the cause of most tax litigation.
There can be little doubt that the courts’ adoption of transplanted doctrines such as the concept of income was thus a root cause of tax complexity. But to blame judges or a handful of transplanted doctrines for ongoing complexity requires a very blinkered view of causation. The judicial doctrines that led to such complexity could have been easily reversed by the legislature simply by substituting more rational statutory definitions if judicial definitions turned out to be inappropriate for tax purposes. Thus, for example, if the judicial concept of income had been identified as a source of complexity, the judicial notion of ‘ordinary’ income could have been displaced by a statutory definition akin to profits in the accounting sense (presuming that any definition of income in an economic sense, including unrealised accrued gains, was a political impossibility).
Focusing on the legacy of transplanted doctrines also shifts attention from the causes of this method of interpreting law. Courts turned to doctrines familiar from other areas of law not only because this was the common law tradition but, importantly, because the legislature adopted provisions that incorporated undefined concepts. Indeed, it could be argued with some force that the legislature deliberately invited the use of transplanted doctrines by the courts. Consider the situation when parliament made tax rules for ‘employees’ without providing a definition of employee in the tax law. Is it likely the legislation would have expected the courts to devise a definition consistent with unarticulated tax law principles when it knew there was a large body of cases available to the courts on the meaning of employee in industrial law and tort law (the former dealing with entitlements to industrial benefits and the latter dealing with the imposition of vicarious liability on employers)?
Similarly, by using the term ‘income’ with no clarifying definition, could the legislature have expected the courts would adopt a concept of income akin to the accounting notion of realised gains or the economic concept of accrued gains when comprehensive judicial doctrines on the meaning of a much narrower income concept were available in trust law? When parliament inserted a measure in the law that provided concessional treatment of gains realised on the sale of small business goodwill (half the gain was exempt from taxation), while offering no guidance on which goodwill qualified and no indication in the exemption section of the object or purpose of the exemption, did it leave courts with any basis for interpretation apart from transplanted doctrines? When parliament moved the concession for goodwill to another measure, again without providing a definition, could it have expected courts to abandon the transplanted doctrines they had adopted to define the term?
In the absence of any source of empirical data, we can only speculate as to whether silence by the legislature genuinely amounted to before-the-fact endorsement of the transplanted concepts it left the judiciary to adopt. It is possible, and perhaps even probable, that had the legislature articulated the boundaries of key concepts more clearly, the result may have largely replicated the judicial interpretations in many respects. For example, in the second decade of this century, when the first Commonwealth income tax was adopted, few beyond a small handful of public finance scholars would have understood why income for tax purposes should in theory include unrealised gains and losses. To the extent the judicial concept of income was restricted to realised gains, it almost certainly would have paralleled the intention of the legislature with respect to the undefined term ‘income’. It is possibly also the case that the legislature would have considered windfall gains on assets that unexpectedly increased in value to be outside the income concept and here, too, the judicial concept may have accurately reflected the concept the legislature had in mind when it used the undefined term ‘income’.
While the basic qualities of judicial concept income may have been intended by the legislature, the government of the day was unlikely to have appreciated the full consequences of leaving definition of the tax base to the courts. For example, the legislature is unlikely to have understood that the judicial concept of income would exclude many ordinary investment gains derived outside the course of regular business. Similarly, the legislature most probably did not expect the judicial concept of income to be so narrow that it would allow taxpayers to easily convert employment income or business income to non-assessable capital gains by presenting it as, say, negative covenant payments or payments for the sale of contractual rights.
Almost every tax concept that derives from an inappropriate transplanted doctrine led to avoidance opportunities and the development of schemes and arrangements to exploit the limits of the transplanted concepts. Each time, the legislature responded to the avoidance with complicated anti-avoidance measures. In the first instance, thus, it is tempting to lay blame for the resulting complexity at the feet of the judges whose doctrines led to the opportunities for avoidance. But an equally logical approach would be to blame the legislature that provided implied invitation to the judiciary to continue to rely on transplanted concepts inappropriate for tax law. In defence of the legislature, in turn, it can be said that without the benefit of hindsight, it is not improbable that the legislature would likely have adopted definitions similar to judicial notions had it chosen to legislate definitions initially. Ultimately, perhaps the blame lies neither with the judiciary nor the legislatures of the time, but with later legislatures that compounded the initial defects in the law first by failing to reverse the initial distinctions when the consequences of transplanted doctrines were revealed and second by later accepting those distinctions and building them into the subsequent statutory framework.
(ii) Colonial Cringe
Although Australia became notionally independent almost 100 years ago, for most of its history the UK-based Privy Council was Australia’s final court of appeal and the Australian common law legal system derives entirely from UK law. One key feature of Australian legal jurisprudence is what has been called the ‘colonial cringe’ factor, a term that signifies the historical preference for ‘mother country’ judgments. To this day, many courts cite UK precedents as readily as Australian ones. The reliance on UK judgments in tax cases is somewhat bizzare given the radically different statutory regime.
The UK income tax law is a schedular law whose ‘schedules’ and subcategories known as ‘cases’ have no legislative counterpart in the Australian global income tax legislation. In fact, Australian legislators deliberately shunned not only the form of the UK legislation, but also its terminology when drafting the first Australian income tax laws. This attempt to start afresh proved to be in vain, however, as Australian judges interpreted the Australian law by applying UK precedents, quite often from relatively low level courts, interpreting completely different terms. Thus, for example, if an English court found that a particular type of payment was not an emolument from office as required by UK income tax law (whether or not it constituted income in a broader economic sense), an Australian judge would likely conclude the gain fell outside the boundaries of the judicial concept of income applied in the Australian legislation.
The subservience of Australian courts to inappropriate English precedents is quite striking, particularly in the period prior to 1980, by which time Australian judicial income tax concepts were largely settled. Rarely did Australian judges point out that a precedent based on an entirely different statutory construct might not be appropriate in Australia and when one did, almost inevitably this view was universally rejected by commentators and other courts. Only in relatively recent times have Australian courts dismissed English decisions as inappropriate and often these cases have involved rejection of base-broadening anti-avoidance doctrines developed by UK courts. The shift of tax jurisprudence from state Supreme Courts to the Federal Court has accelerated this process as Federal Court judges have distanced their jurisprudence from UK precedents, but of course the earlier Australian precedents they cite in preference to UK cases were themselves often derived directly from UK decisions.
By narrowing the judicial concept of income, colonial cringe decisions contributed to the fragmentation of the income tax base. Once again, however, the real complexity arises not so much from the initial formulation of a narrow and fragmented income tax base, but rather from the legislative response to the original decisions. Sometimes the legislature reversed the effect of colonial cringe cases with narrow statutory inclusion measures. But often, remarkably, the legislature showed the same deference to English courts as the Australian judiciary, and actually legislated to insert perverse English base-narrowing decisions into the statute, ignoring the fact the UK legislature had subsequently overturned the decisions explicitly by legislation.
A classic example of this behaviour is the characterisation of so-called ‘negative covenant’ remuneration paid for an employee’s or service provider’s undertaking not to compete with the employer or service user. When UK courts decided this type of gain fell outside the narrow schedular remuneration rules in that country, Australian courts concluded the gains must fall outside the Australian judicial concept of ‘ordinary’ income. The lone voice of a High Court judge questioning the basis for this approach was generally ignored  and UK doctrine helped define a key concept in Australian law.
The Commonwealth Parliament made no change to provisions affecting remuneration to overcome the decision. Many years later when legislation was introduced to bring retirement payments into the tax base, the legislation contained an explicit exemption for negative covenant payments. A few years later, when a comprehensive fringe benefits tax was adopted, the fringe benefits rules also included an explicit exemption for negative covenant payments. In both instances, the legislature accepted the judicial characterisation of these gains as ‘capital’ gains outside the income concept. When the gains were finally brought into the tax base, they were not identified as a type of income but rather were deemed to be capital gains, assessable under the capital gains rules, first by means of complex artificial deeming provisions and later through an equally artificial ‘capital gains event’.
Rather than articulate clear boundaries to concepts such as income, the legislature responded to unacceptable colonial cringe decisions with piecemeal, ad hoc measures intended to overturn particular judgments, and sometimes, as has been seen, by first endorsing the decisions. Instead of broadening the tax base, the inclusion measures, when they came, often had the effect of narrowing it. By spelling out in great detail situations in which gains constituted income assessable under Australian law, the sections implied, to the judiciary at least, that amounts slipping outside the literal boundaries of the new provisions were not to be treated as assessable income. Time and again a decision would trigger a narrow inclusion measure which in turn would be misapplied by the courts to tighten the base further, then prompting another round of limited inclusion measures and so fuelling a vicious cycle.
(iii) Strict Literalism
The debate between those advocating purposive interpretation and those who support literalist interpretation of tax law is not confined to Australia; long before the Duke of Westminster sought to avoid tax on income used to pay his gardener, English courts and their counterparts throughout the English-speaking world debated the proper mode of statutory interpretation. On one side of the debate was the purposive school, whose adherents argued that to determine the tax consequences of a transaction courts should look through its superficial legal form to its underlying economic substance. On the other side was the literalist school, whose members argued courts should respect the legal form of an arrangement and look no further. This school argued that looking beyond legal form amounted to an inappropriate usurpation of legislative function by courts.
In the eyes of many, the apogee of the literalist school in Australia was reached in the late 1970s and early 1980s under the tutelage of Sir Garfield Barwick, Chief Justice of the High Court. During the period of Sir Garfield’s leadership, the High Court condoned a range of tax minimisation arrangements including income splitting devices, transactions to strip company profits free of tax, transfer pricing schemes to shift profits to low tax jurisdictions, schemes to convert otherwise taxable income into exempt foreign-source income, and many more. Some speculated that the apparent consistent judicial endorsement of tax avoidance was ideologically motivated, reflecting the preference of the Chief Justice for individuals over the state or personal wealth over public accumulation. Others strongly dispute this view, but it must be conceded that often the Chief Justice’s opinions contain thinly veiled admiration for the clever way in which taxpayers had manipulated inconsistencies in the tax law to further a tax avoidance scheme.
There is no doubting the nexus between literalist interpretation and complex tax legislation. To begin with, the legislative drafters, wary of judicial dismemberment of their work, strive to cover every possibility and include every eventuality in provisions, with unwieldy, complex legislation the inevitable outcome. Then, when literalist interpretation does lead to avoidance, the drafters’ response is inevitably, but understandably, excessive, adding new complexities to the proven ineffective older complexity. But while there may be a direct cause and effect relationship between literalist interpretation and complex legislation, both may merely be symptoms of a deeper, and more pervasive, complexity in the tax system.
It is quite true that avoidance schemes relying on a sympathetic literalist hearing are premised on the hope that a court will accept the superficial form of a transaction and not seek to ascertain its true economic nature. But does it necessarily follow that a court which looks through the transaction to its underlying substance is necessarily protecting the legislative intent?
The issue is somewhat complicated by the judges’ distinction between legal substance and economic substance. This difference is well illustrated by a more recent scheme of the early 1990s in which a bank recast a standard credit foncier loan agreement as an annuity to take advantage of the deferred recognition of income available to holders of annuity contracts. Historically, annuities were blended payment loans to life insurance companies where the return contained a mortality bet, with payments on life annuities based on the investor’s expected life rather than a pre-determined payment period. Because the actual period of payments was not known at the start of the repayment period, the conventional formula for calculating the interest and principal repayment portion of each payment could not be used. The tax law allowed taxpayers to recover the amount invested as repayments of principal on an equal basis over the expected life of the loan based on mortality tables, effectively deferring recognition of some of the interest component from the early years of the annuity to the latter years for significant tax savings.
When financial institutions began to offer fixed term ‘annuities’, the government continued to allow taxpayers to use the annuity principal recovery rules even though the rationale for the special treatment no longer applied. Not surprisingly, aggressive taxpayers sought to recast ordinary loans as annuities so they could defer recognition of the interest they derived through the loan now called an annuity. The leading case involved a large bank loan to a state government institution which had agreed to the rebadging of a standard loan agreement on the promise that it would enjoy a share of the tax savings by way of lower interest charges. At trial level, the presiding judge applied a purposive approach to characterising the transaction and treated the payments received by the taxpayer as payments received on a loan, concluding the transaction was simply a loan rebadged on paper as an annuity. The appeal court, however, accepted the form of the transaction, respecting the title of the document.
Did the former judge protect the legislative intent behind the law more than the latter judges? Arguably not. While there is no difference in terms of economic substance between a credit foncier loan and a fixed-term annuity, there is a difference in terms of ‘legal’ substance. The common law phrases the obligations and rights of the parties to the two different types of contracts in different terms and draws upon legal concepts from very different historical sources to describe the legal relationships between the parties to each type of transaction. Although the net economic effect of the arrangements may be identical — they both constitute an obligation to repay over time by means of regular equal payments the principal deposited and a return for the use of capital — the so-called legal substance of the two arrangements is conceptually quite different. The appeal court in the bank loan/annuity case simply respected the legal substance of the transaction.
By employing appropriate language, a lender can with relative ease draft a contract offering the same economic benefits as either an annuity or a credit foncier loan. A rational lender, aware that distinctions rooted in century old English judicial concepts had been accepted and codified into tax law by the Australian Parliament and aware that dramatically different tax consequences could turn on the distinctions, would be foolish not to structure a loan in the tax-preferred legal form. In the absence of any policy basis for the incorporation into tax law of an economically irrelevant legal distinction, it would be difficult indeed to claim judges were amiss in characterising a transaction by reference to superficial legal form rather than underlying economic substance.
To assert that in most cases of strict literalism the courts are simply allowing taxpayers to exercise a choice to which the legislature had explicitly or implicitly acceded is to ignore, however, the many instances in which courts have frustrated the logic of an income tax, if not the strict words of the law. A leading and recent example is the High Court’s treatment of a reimbursed expense. An income tax should tax net gains, not gross receipts, and accordingly, the legislation allows deductions for expenses incurred to derive gross income. If a taxpayer subsequently is reimbursed the cost of a previously deducted expense, the deduction should be reversed and the most obvious way to do this is to include the reimbursement in assessable income. It is obvious to both tax policy experts and lay persons that this approach is consistent with the logic of the income tax and it is not surprising that outside Australia there are courts that have readily endorsed this principle.
When the issue was presented to the High Court as a matter of principle, however, the High Court recoiled. The judicial (ordinary) concept of income is based not on logic, the Court reasoned, but on a long line of cases setting out criteria based on the characteristics of a receipt and a perceived nexus with income-producing activity. The Court very deliberately rejected the proposition that its view of income should in any way be influenced by the clear logic of the income tax. Even the minority that found the reimbursement was judicial concept income rejected a place in Australian tax jurisprudence for the logic of a recoupment principle, instead finding the reimbursement constituted ‘ordinary’ (that is, judicial concept) income because it enjoyed a nexus with the taxpayer’s services to his employer.
It is thus possible to excuse many, if not all, exercises in strict literalism as a rational reaction to legislative ambiguity or choice. To this extent, it is difficult to blame the courts for the complexity to which they react. At the same time, however, there is a long line of cases in which the courts have directly contributed to tax complexity, originally by slavish adoption of inappropriate transplanted doctrines and UK precedents, and continuing with blind adherence to traditional doctrines that appear to undermine the logic of the income tax.
Tax accountants and lawyers are commonly blamed for many of the ills of the tax system, and in particular its complexity. The flood of cases annually heard in the tax tribunals and courts, and the aggressive marketing of tax minimisation investments, are seen by many as evidence of the fact that tax advisers are behind most avoidance activities in Australia and thus responsible for the complexity to which the legislative reaction inevitably leads.
There is no doubt that Australian tax advisers devote considerable effort to devising tax minimisation schemes on behalf of their clients. While neither taxpayers nor their advisers are as blatant and brash in their pursuit of tax minimisation as they were in the heyday of tax avoidance in the late 1970s and early 1980s, when avoidance and evasion were estimated to cost Australia billions of dollars in lost revenue, the level of aggressive tax planning and scheme marketing seems to have abated only slightly since.
In the heyday of scheme marketing, critics sometimes argued that tax advisers who devised avoidance schemes were failing in their social and professional responsibility. In a well-publicised exchange between tax lawyer Mark Leibler and legal academic Dr Yuri Grbich, Mr Leibler argued that quite to the contrary, this behaviour was entirely consistent with the duty of a lawyer. It was the responsibility of the legislature drafting tax law to prevent avoidance, he argued, not the lawyer whose social responsibility was fulfilled by acting on behalf of the client.
While critics of tax lawyers may be tempted to dismiss Mr Leibler’s views as a rationalisation for the role of tax advisers in the world of tax minimisation, a closer look at the tax avoidance cases reveals the logic of his position. The common feature of every so-called tax avoidance case is the exploitation of a legal distinction that has no counterpart in a neutral and equitable benchmark income tax. For example, a large number of cases in the 1970s involved taxpayers seeking to recharacterise private companies so they fell within the definition of a public company. This was done because the tax law sought to impose annually two levels of tax on profits derived through private companies while allowing investors in public companies to defer the second level of tax indefinitely. There was, however, no sharp sustainable distinction between the two types of company explicable by reference to some underlying tax policy.
In other cases, taxpayers sought to extract company profits by way of a sale of shares in a company rather than directly as dividends, since gains on the sale of shares were completely exempt from tax while dividends were fully taxed, a differential treatment with no tax policy basis. When the tax law made a distinction between gains on shares sold less than a year after acquisition (which were fully taxed) and those made on shares sold a year or more after acquisition (which were completely tax exempt), taxpayers owning both types of shares in a company would arrange for the company to modify its constitution so all value was shifted from recently acquired shares to the ones held for more than a year. When the tax legislation imposed full tax on domestic source interest and exempted interest derived offshore provided it was subject to nominal tax even in a tax haven, taxpayers devised paper transactions to shift the source of income from domestic banks that actually guaranteed the debt to notional deposits offshore. When the tax law imposed high tax on income diverted to a child via a trust established directly by the child’s parent, parents would arrange for a friend or accountant to establish the trust and then deposit an amount in the already established trust. When the tax law required upfront recognition of the interest component of a blended payment loan but allowed deferred recognition of gain if the loan was structured as an annuity, taxpayers opted to call their loans annuities. Examples are legion.
Inconsistent rules in income tax offer taxpayers choices and it is not surprising that taxpayers faced with an initial choice would choose a form of transaction offering the lowest tax burden. It is similarly not unexpected that taxpayers finding themselves on a heavily taxed side of a distinction would ask their advisers to devise schemes that moved the form of their investment or the legal nature of the transaction into a less heavily taxed arrangement. An adviser who did so by dishonestly portraying a transaction as something other than what it was — signing off a tax return indicating a company was a trust or vice-versa — would clearly be in breach of ethical and moral professional standards (not to omit committing fraud). But is the tax adviser who redrafts the constituent documents for an entity to convert it from a company to a trust or rephrases a loan agreement to convert it to an annuity seeking to frustrate deliberately the intention of the income tax law?
In the many cases in which tax minimisation schemes were successful, advisers had successfully converted or recharacterised transactions from a legal perspective. To suggest this action was contrary to the intention of the income tax law would require an assumption that rational tax policy was built on the underlying economic substance of a transaction rather than its legal form and it was thus somehow an abuse of legal manipulation to disguise one type of economic substance as another legal form.
This argument is impossible to sustain in almost every instance of tax minimisation. The difference between a company and trust is only one of legal form — properly drafted, any trust agreement can replicate the commercial structure of a company and vice-versa. The only difference between a withdrawal of company profits by way of direct dividends and by way of a capital gain with the purchaser declaring the dividend to fund the purchase is an intervening legal contract with no ultimately different economic effect. The only difference between a sale 364 days after acquisition and one 366 days after is an arbitrary legal one. The only difference between domestic source income and offshore income is the legal source — the economic income itself is identical in both cases. The list goes on and on.
In not one of the tax minimisation schemes developed or implemented by tax advisers was there a rational underlying economic basis for the distinction in the tax law that prompted the schemes in the first place. It may well be true that in each case the tax advisers took a transaction, entity or arrangement that legally appeared to fall on one side of the line and transmogrified it so that it legally fell on the other, but it is difficult to find any case in which the underlying economic substance of the transaction changed one way or the other. In other words, the tax consequences depended not on the substance of the transactions but their superficial legal form. There is no basis for blaming the tax advisers for their role in minimisation if the legislature decides to erect a tax system built upon the fragile facade of legal forms and taxpayers engage tax advisers to minimise their taxes by manipulating legal forms. It would be naive at best to think a tax law based on legal distinctions with little or no underlying substance would not invite taxpayers to recast the form of transactions, and it is fanciful to think tax advisers assisting taxpayers to cross such artificial thresholds are the persons ultimately responsible for tax avoidance. The culprit is the legislature that introduced the thresholds, not the taxpayers who seek to cross them or the tax advisers whose job it is to assist taxpayers to achieve economic benefits.
Of all the parties who contribute to the development of tax law in Australia, the legislative drafters appear to be the most difficult to paint as primary culprits. True, they are responsible for turning out the complexity inherent in the legislation. But the drafters are mere servants of the revenue officials who provide drafting instructions and to blame the drafters for faithfully implementing the directions handed to them at first glance seems to be an instance of blaming the messenger. Closer examination, and in particular recent changes in Australian tax drafting techniques, however, suggests there may be some basis for criticising the drafters.
Until the late 1990s, Australian law, including tax law, followed the British drafting tradition of single sentence sections, without regard to the number of subsections, paragraphs, subparagraphs, sub-subparagraphs and sub-sub-subparagraphs that might be contained within a section and without regard to the number of exceptions and provisos that led to double, triple or even more negatives in a section. With provisions covering many pages, it is not surprising that the law is little better than unreadable gibberish at times. It is probable that many tax judges quietly share the view expressed by Hill J in a recent case: ‘While it is for the courts to endeavour, where possible, to give effect to the legislative purpose, it should not be expected that the courts will construe legislation to make up for drafting deficiencies which revel in obscurity.’
In the latter half of the 1990s the Australian Taxation Office embarked upon a major project to redraft parts of the income tax law in simpler, ‘plain English’. As is explained in more detail below in section 3A, when well less than half the law had been redrafted the project was subsumed into a review of business taxation and it was never officially revived when that finished. Still, since 1997 all new tax legislation is supposed to have been drafted using ‘plain English’ construction to simplify the law. The law nevertheless has grown in complexity. One undeniable factor is bad drafting. Notwithstanding the lofty goals of the plain English simplification project, the drafters continue to turn out measures with constructions and meanings that are counter-intuitive to the point of obscurity. It often appears as though their object is to deliberately mystify by means of intricately designed mazes erected as barriers to prevent all but those who designed the measures from understanding them.
A nice illustration of the problem is concessional provisions adopted in 1999 to halve the tax levied on most types of capital gains. With a disregard bordering on disdain for the principal of plain English drafting, the designers of the new concessions tortured the English language into constructions that defy any logic and displaced ordinary meanings with completely counter-intuitive terms and constructions. The capital gains concessions revolve around a concept known as a ‘discount capital gain’. In ordinary usage, the adjective ‘discount’ means a figure that is reduced from a non-discounted figure — a discount price, for example, is less than the ordinary retail price. But in the capital gains tax legislation, ‘discount’ is given a meaning precisely the opposite of its ordinary meaning. A discount capital gain is the full amount of a capital gain that may potentially be discounted if qualifying tests are satisfied.
In section 3A below, the question whether changes in drafting technique alone can simplify tax law is considered. Even if it cannot (the discussion below suggests that is the case), it need not complicate it. But this observation appears to have escaped the attention of Australian tax law drafters who fail to see a problem with the use of convoluted provisions and artificial definitions so that the statutory meaning of words is sometimes completely the opposite of ordinary usage.
Commonwealth income taxation has been in effect in Australia for close to nine decades and it is probably a fairly safe generalisation to assert that Treasury officials will rarely encounter genuinely new tax issues or propose truly new policy initiatives. There are, of course, exceptions to the rule — in the mid-1980s Australia moved from a classical company and shareholder income tax system to an imputation system, shifted fringe benefits taxation from employees to employers and, following the dismantling of foreign exchange controls, adopted a comprehensive foreign tax credit system, followed after a few years by the adoption of four attribution regimes to prevent offshore deferral through companies and trusts. By and large, however, Treasury has enjoyed several decades in which it should have been in the position of refining tax laws rather than designing de novo ones (apart from devising the inevitable tax expenditures demanded by governments from both sides of the political spectrum).
Treasury’s refining responsibilities have largely been responsive in nature, often dealing with loopholes caused by undesirable judicial decisions traceable to one or more of the problems of doctrine misappropriation, precedent misapplication, or policy-defeating strict literalism. In virtually every case, the government’s response was the wrong one and a legitimate question to ask is to what extent should Treasury bear responsibility for bad laws. On the one hand, Treasury can rightly say it is merely responsible for carrying out the policy of the government of the day. At the same time, however, in its capacity as adviser to the government, Treasury has a responsibility to protect the integrity of the revenue system and accommodate the wishes of the government of the day in a manner that establishes and reinforces that integrity rather than undermining it.
There are three possible explanations for inadequate or inappropriate recommendations in response to emerging tax problems. One is that Treasury did not actually understand the fundamental tax principles that should have been guiding its advice. If Treasury understood basic timing recognition rules, would Australia have ended up with 36 amortisation regimes for capital expenses when one would have sufficed?  A second possibility is that Treasury understood the principles but opted for recommendations of ad hoc inclusion, exclusion or shifted timing rules as the easy solution to immediate problems. To use the amortisation example again, once the pattern of ad hoc add-ons had started, was it not expeditious to simply add more in preference to revisiting the basic rules? And the third possibility is that Treasury understood the principles, framed its proposals in appropriate structural terms and its advice was repeatedly rejected by governments which suggested ad hoc solutions in preference to the approaches recommended by Treasury.
The first possibility is not out of the question. Treasuries overseas often establish permanent tax policy divisions to retain and nurture tax policy expertise. By way of contrast, the Australian Treasury tends to rotate its personnel through its functional divisions, inhibiting, if not preventing, the development of mature tax policy expertise, not to mention the facility to build and pass on institutional wisdom. Also, until very recently, Treasury had little or no genuine technical tax expertise. Its personnel may have had some exposure to general public finance theory, but unlike many of its counterparts abroad, it had no tax lawyers or accountants on staff and thus had little or no idea how transactions might be structured in the real world. It had to rely on advice from the Australian Taxation Office on how best to implement its recommendations and almost without fail the ATO officers who provided the advice had no expertise in tax policy principles.
The third possibility is also conceivable, though highly unlikely. Governments are interested in outcomes, not the details of the legislation used to achieve those outcomes. To be sure, Treasury was often advising governments whose political or ideological agenda quite definitely rejected comprehensive reform of the sort truly needed to solve the problems to which the government was forced to respond. However, within this constraint Treasury could have rigorously pursued options that would minimise the damage caused by the fractured tax base. If Treasury had consistently devised responses in structural terms — carving exceptions out of a broader base rather than adding or taking away from a narrower one — governments would most likely have accepted Treasury’s advice so long as their immediate tax programs were achieved.
The second possibility emerges as the most likely explanation, albeit facilitated in many instances by an inadequate understanding of policy principles. There has of late been some recognition of the technical shortcomings of Treasury and the failure of Tax Office advisers to appreciate the broader policy principles. The government moved to address these two issues in 2002 by moving the legislative design branch of the ATO into the Treasury. It is too soon to say whether the changes will ultimately result in a shift towards policy-based recommendations for legislative design. For the moment, ad hoc responses remain the norm.
A fair assessment of the legislature’s contributions to the complexity of Australian income taxation requires dissection of the legislature’s role into two components, the first being the response of the legislature to judicial decisions or tax schemes that threatened the revenue or integrity of the income tax system and the second being the legislature’s predilection for using tax laws as a tool of economic and political intervention.
(i) Responding to the Judiciary
At first glance, a review of complexity resulting from inappropriate or piecemeal legislative responses to judicial decisions that open holes in the revenue base might suggest that the legislature is a wrongly accused party. In each case the government was informed by revenue officials or Treasury officers of a problem and it reacted through amending legislation. The laws in question were reviewed and designed by Treasury and drafted by parliamentary counsel based on detailed drafting instructions provided by the Australian Taxation Office. If the resulting tax law was overly complex because Treasury recommended a piecemeal ineffective response rather than a comprehensive structural response, is the legislature, which necessarily relied on the persons preparing the law, a legitimate target for blame?
The answer in many cases is an unequivocal yes. It is quite true that time and again Treasury handed the government ad hoc and piecemeal responses rather than better formulated structural solutions. But it is also the case that Treasury deliberately failed to push the government for comprehensive solutions because it recognised that the government’s apparent complacency with respect to the underlying problems often masked an ideological commitment to retention of the underlying structural flaws or concern over electoral risks inherent in three-year terms of office.
The phenomenon is well illustrated by the Coalition Government’s reaction to an era of unprecedented avoidance and evasion in the late 1970s and early 1980s when aggressive tax planners devised a multitude of avoidance schemes that exploited the judicial distinction between income and capital gains. The government knew full well that broadening the income tax base to include capital gains as defined by the courts was the only effective response to the schemes (and could potentially significantly reduce the complexity of the tax law). But it directed Treasury to devise narrow responses to particularly costly schemes, a policy that led to what is undoubtedly some of the longest, most difficult to understand, and most complex anti-avoidance provisions in the law. It would appear the government of the day had a strong ideological commitment to a narrow tax base falling excessively on labour income, being interested in neither equity nor efficiency as benchmarks for the income tax base. It seemed that the government’s immediate objective was to retain the exemption for many or most types of gains realised by wealthy individuals. It was also strongly committed to retaining an exemption for capital gains derived by corporate taxpayers.
Faced with such ideological intransigence, Treasury could only advise the government on narrow measures targeted at the most costly and abusive schemes. The deliberate rejection by the conservative Fraser government under Treasurer John Howard (1975–1982) of reforms that might involve base-broadening directed at high wealth individuals and corporations exploiting the judicial income–capital dichotomy is but one of many examples of governments quite deliberately choosing complexity over simpler, but politically objectionable, systematic solutions.
(ii) Using Tax Law to Achieve Social and Political Objectives
Australian governments’ deliberate proactive intervention in the economy through tax legislation is another prime reason for complexity in the income tax system. Notwithstanding the rhetoric from all sides of the political spectrum that in theory Australia favours a market economy as the vehicle for optimal distribution of economic resources, in practice governments of all shades appear to have little faith in the ability of the unassisted market to deliver optimal economic and employment growth. To correct perceived market failures or in the belief or hope that they are better than the unfettered market at picking winners, governments regularly intervene in the market with a range of subsidy programs to alter the allocation of investment and consumption. Added to the economic aims of subsidies is the inevitable objective of seeking friends and winning votes in marginal seats through corporate and individual welfare handouts.
Increasingly, Australian governments have relied on tax expenditures in preference to direct expenditures to carry out their subsidy programs. Tax expenditures are indirect spending programs that deliver benefits by way of remissions of tax that would otherwise be payable in the absence of concessional subsidy measures. They are accounted for in an annual tax expenditure budget but because they do not require annual appropriation bills as do direct expenditures, handouts through the tax system are far less transparent and far more immune from public scrutiny than direct expenditures. This is seen as their most positive feature by governments and the beneficiaries of tax-based subsidies since often the recipients of tax expenditures adopt public positions opposing increased government spending or subsidy programs.
Since the early 1970s when Stanley Surrey penned Pathways to Tax Reform, western governments have acknowledged the inefficiencies and inequities inherent in tax expenditures and have equally acknowledged the benefits of direct expenditures. But recent Australian governments have favoured the income tax law as a convenient spending vehicle and the tax expenditures embedded in the Income Tax Act now far outweigh the actual income tax measures in terms of number of provisions, if not yet in monetary effect. Tax expenditures are growing annually and now equal one-fifth of total income tax collected, accounting for one-eighth of total government expenditures.
Australian taxpayers receive tax subsidies for investments ranging from Australian films to vineyards, and for activities ranging from research and development, through the provision of on-site child care for workers, to water conservation. Accelerated cost recovery measures for many categories of assets seek to divert investment away from some sectors and to others. Generous valuation rules subsidise the provision of automobiles in remuneration packages. A full portfolio of subsidies is provided to small business owners (or at least to those who sell out of successful small businesses). Subsidies are provided for a number of different types of savings, particularly retirement savings. The list goes on and on.
Using, or more accurately, misusing, the income tax law as a spending vehicle is undoubtedly one of the largest sources of complexity in the legislation. It has proved impossible to deliberately distort investment or consumption behaviour by lowering the tax burden on preferred activities and not invite abuse. Tax law never specifies the intended recipients of concessions; at best it seeks to define the types of transactions or investments that will qualify for tax expenditures. However tightly the boundaries of desired activities and assets are defined, it is inevitable that they will be breached by well advised taxpayers recharacterising transactions and investments to qualify for the subsidies. This activity, in turn, will lead to complex anti-avoidance measures intended to protect the integrity of the original subsidy scheme. The new legislation will lead to further planning which will lead to further legislation, and the cycle will continue for many years until either the concession is abandoned or is buried within dozens of complex anti-avoidance provisions.
The grim picture just painted is in fact a best case scenario. In many cases tax expenditures are adopted with no attempt to define or delineate the activities or investments that will enjoy concessional treatment. The leading example is the tax subsidy for capital gains. To encourage investment in riskier though potentially more rewarding assets, the government provided a number of preferences for capital gains when it finally included judicial concept capital gains in the tax base, but never inserted a definition of the gains that should qualify for the concessions.
The consequences of subsidies for ill-defined beneficiaries with no articulated or apparent policy objectives behind the subsidies are illustrated well with the small business concessions.
(iii) Small Business — A Case Study
Without doubt, no sector of the Australian economy enjoys more subsidies through the tax law than the small business sector. While there is little or no empirical evidence to support the case for subsidising the small business sector, there are obvious political gains to be made from support for the sector and over the years a wide range of tax subsidies has been adopted for small business. Among other things, small business or owners of small business may qualify for:
In 2001, the government adopted new tax concessions for small business, presented as a ‘Simplified Tax System’. The ‘simplification’ measures provide a new optional depreciation regime for qualifying small businesses (with two concurrent mandatory depreciation systems for particular categories of assets), new cash basis accounting rules for qualifying small businesses, and new trading stock accounting rules, again for qualifying small businesses. The legislation suggests that in usual circumstances the rules will simplify the calculation of taxable income and reduce compliance costs for qualifying taxpayers.
This assumption is questionable. Since the tax rules adopted differ substantially from accounting rules, many small businesses that use the new tax rules will have greatly increased accounting costs as financial accountants maintain multiple (and different) depreciation schedules for tax purposes in addition to the ordinary accounting depreciation schedules and establish different trading stock accounts for accounting and tax purposes. Those who want to use the concession exempting them from stock valuations would be well advised to first carry out a stocktake to confirm that their closing stock on hand falls below the threshold for which stock valuation is necessary. Turnover will have to be rigorously reviewed to ensure the qualifying turnover threshold for access to the concession rules generally is not breached, and relationships with associated businesses must be monitored to ensure all required turnovers are included in turnover calculations. After all this, many small business taxpayers will calculate taxable income regularly first using the ordinary rules and then using the Simplified Tax System rules to see which yields the best tax result for them in any given income year. In short, the ‘Simplified Tax System’ adds a new set of tax rules onto the existing rules and for many taxpayers will require double the work, an odd understanding of tax ‘simplification’.
The Simplified Tax System rules illustrate well how tax concessions, even those allegedly adopted to simplify tax compliance, inevitably led to more complication in the law itself, as well as its application. The legislation itself adopts a peculiar structure for allegedly simple law — only after several dozen pages describing the concessions does a reader finally reach the provisions that explain whether any of the preceding pages have any relevance for a given taxpayer — that is, whether the taxpayer qualifies for the concessions. Eligibility turns on a variation of business turnover called ‘STS turnover’, subject to a cap on the value of depreciable assets owned by the business. It was expected that to avoid the caps and qualify for the generous accelerated depreciation that the rules allow, taxpayers would split investments or activities between different entities and comprehensive rules consolidating turnover of affiliates were adopted to prevent this. This in turn required comprehensive definitions of affiliated and related entities. The net result is that the anti-avoidance measures in the Simplified Tax System measures account for a good percentage of the law as drafted. Despite the pretentious title, the Simplified Tax System concessions provide yet another example of how providing subsidies through tax expenditures is an inevitable recipe for complexity in tax law.
While there may be little agreement among the various players in the tax game as to who is to blame for complexity in the income tax or, for that matter, exactly what simplicity entails, there is near universal agreement that ‘simpler’ law of some sort is needed. In the late 1990s two new models for tax simplification were unveiled to the Australian community — the first a proposal to rewrite the entire law in ‘plain English’ and the second a grandiose plan to rewrite the law upon a new foundation, labelled the ‘tax value method’, substituting for the framework principles that had guided the income tax system for the past eight and a half decades.
The idea of wholesale replacement of the current law with a new one was not new. It was a path strongly advocated in the early 1990s by Sir Harry Gibbs, former Chief Justice of the High Court, who predicted the first approach, simplification by way of redrafting, would fail: ‘[The Act’s] complexities cannot be removed simply by rewriting the existing provisions in plainer language .... Real reform would require not a rewriting of the present law, but a completely new Act; that is, a new approach is necessary, and to put the existing provisions into new words would not by itself be enough.’
The plain English redraft proposal was strongly supported by the Australian Taxation Office as the solution to complexity and when work on that proposal stalled, the tax value method proposal was enthusiastically supported by both the Australian Taxation Office and the Treasury as the preferable model for tax simplification. The former was initially strongly resisted by tax advisers but later accepted by most when they realised the changes to the law with which they regularly worked were at best superficial. The latter proposals were rejected almost entirely by the business community and tax advisers.
The first failed model provides a useful insight into the inadequacy of reforms based on superficial change while the latter illustrates well how not to pursue simplification in a world of tax Realpolitik. An analysis of both sets of proposals is useful to set the stage for a more realistic simplification program.
Simplification through drafting alone was first proposed in 1990, by the then Treasurer, Paul Keating, who appointed a joint Treasury and Australian Taxation Office task force to investigate the idea. The simplification team concluded that there was little to be gained by rewriting the law until tax policy was simplified. This advice was politically unpalatable to the government and the project’s consultative paper was never released.
The simplification through drafting project was revived in 1993 following the release of a parliamentary committee report that implied simplification could be achieved by means of drafting changes alone. This advice, complementing the government’s ‘no policy change’ agenda, was welcomed by the Treasurer, who announced a major simplification project to redraft the income tax law into ‘plain English’. The goal of simplification was to replace the layers of obscurity and uncertainty in the law with simple, clear and unambiguous provisions.
The re-draft project, accorded the grand title of the ‘Tax Law Improvement Project’ and commonly referred to in somewhat redundant terminology as the ‘TLIP project’, commenced with a comprehensive review of communications and writing theory before applying the theory to drafting. The result of this effort was a truly revolutionary new style for writing legislation. Traditionalists were at first shocked at the radical new format of early TLIP drafts. The new law was not only written in plain English style, but incorporated guides to the structure of the law, notes with illustrations of the application of sections, and cross-references to other sections that might override or modify the application of the section at hand. It even contained diagrams and flow charts to assist readers in working their way through the law.
For the most part, provisions were drafted in ordinary sentences and almost all the illogical constraints of traditional drafting were abandoned. Also assisting with the readability of the new style was the adoption of a new bifurcated section numbering system with the first half of the number identifying the Division in which a section was located. Since each Division dealt with a discrete topic, the subject matter of a section could be identified just by its number. Divisions and sections were well spaced out with unallocated numbers held in reserve to avoid the problems encountered in the predecessor legislation trying to squeeze new sections between existing provisions.
The government opted for a progressive implementation of the re-written law, despite widespread resistance by the tax profession which called for completion of the rewrite and implementation as a ‘big bang’ project. The idea was that the plain English version of the law would gradually replace the existing tax Act, the Income Tax Assessment Act 1936 (first enacted in 1936 and amended continually since then) as provisions were added to the new Act and repealed in the old one. An initial tranche of the new law was enacted in 1997 (as the Income Tax Assessment Act 1997) and a second tranche was added to the 1997 law in 1998. Parallel with this process, new divisions were appended at the end of the 1936 Act as plain English ‘schedules’, to be inserted into the new law when the appropriate framework was completed.
A superficial look at the 1997 Act supported to some extent those critics who blamed the drafters for the complexity in the old tax law. It appeared the new plain English law was easier to read and comprehend than its predecessor (though a number of commentators noted that the improvement was not terribly obvious in many cases). The improvement led many to conclude initially that the obscurity of the law was the prime culprit behind its complexity, thereby implicating the original drafters. However, when taxpayers and tax advisers sat down to work with the new law, they quickly discovered none of the complexity had dissipated. Indeed, by unveiling many of the inconsistencies, anomalies, overlaps and lacunae in the law, the plain English redraft exposed the real causes of much of the former law’s complexity, namely its wholly irrational and inconsistent policy base. And somewhat ironically, the diversion of drafting resources to the TLIP project was alleged by some to have led to an increase in problems with other tax law drafted by ‘business-as-usual’ teams outside the project.
The limitations of the project are perhaps best illustrated through one of the first areas tackled by the simplification team, the redraft of depreciation and amortisation rules, and one of the last areas addressed before the project was sidelined, the capital gains rules.
(i) Depreciation and Amortisation
The judicial reliance on trust law doctrines that played such a key role in the development of the income concept was paralleled by similar reliance on trust law doctrines to distinguish revenue expenses and capital expenses. The former were charged against the interests of income beneficiaries to a trust while the latter were charged against the interests of capital beneficiaries. The initial income tax law adopted this judicial distinction and provided an immediate deduction for revenue expenses while denying any recognition of capital outgoings unless they qualified for deduction under an amortisation regime.
Only one amortisation regime was included in the original Act, a depreciation regime for expenses incurred to acquire ‘plant and equipment’. These words were subsequently interpreted to mean only machinery, tools, and similar assets so expenses for many types of wasting assets including intangible property and buildings were excluded from the amortisation regime. As a result, expenses incurred to acquire other long-life assets were not recognised for tax purposes. As might be expected, taxpayers went to great lengths to portray otherwise excluded types of assets as plant and equipment.
This regime based on judicial characterisation created a sharp distinction between revenue expenses, always deductible regardless of the length of the benefit to which they related, and capital expenses, never recognised for tax purposes unless they fell into the statutory amortisation regime. Expenses excluded from recognition were once known as ‘nothings’ as they were treated as nothing for tax purposes and later were labelled ‘black hole’ expenses since they fell into an income tax black hole.
Over the decades taxpayers lobbied for expansion of the depreciation regime to cover the wide range of wasting assets used in business. More often than not the legislature acceded to the requests. However, rather than recognise depreciation as an appropriate policy response to the acquisition of wasting assets generally, the legislative designers treated each recognition of a capital expense as a tax concession, adopting separate recognition rules every time another lobby group convinced the government of the day that their specific expenditure should be depreciable.
The plain English redraft of the income tax law in 1997 somewhat consolidated the many depreciation and amortisation rules in the law into half a dozen different places. It did not change the rules — it merely redrafted them into ‘simpler’ English. The foundation of the new draft was a set of ‘common rules’ that, given the range of different treatments accorded wasting tangible and intangible assets, actually applied in few, if any, cases. The remainder of the provisions set out the actual depreciation rules, all exceptions to the common rules.
While the absurdity of multiple amortisation regimes (they number more than 30) was difficult to perceive when they were scattered throughout the legislation, when the multiple regimes were brought together into a more limited number of places, their complexity was readily visible to any observer. Each regime contained different fundamental definitions, different valuation rules, different balancing rules on disposal of depreciated property, and different depreciation formulae. Not only were there different regimes for different types of wasting property, but within each category there were different regimes for subcategories of assets. Depending on the use to which they were put, income-earning buildings might, for example, be depreciated under a regime for mining industry buildings, a regime for timber industry buildings, a regime for hotels and motels, a regime for commercial buildings and residential rental buildings, and so forth.
The inconsistencies, overlap and lacunae revealed by the 1997 partial consolidation of depreciation and amortisation rules led to almost immediate pressure for reform. In particular, the many categories of ‘black hole’ expenses omitted from the depreciation rules featured prominently in submissions to a review of business taxation established by the government in the late 1990s.
The 1997 plain English rules on depreciation were amended many times in the three years following their adoption. Finally, in 2001 the government indirectly conceded that the so-called simplified rules were not working and scrapped the lot, replacing them with a new plain English depreciation and amortisation regime. The new rules provided some further consolidation and provided new, and separate, amortisation rules for a limited number of black hole expenses. They did not, however, address any of the fundamental flaws or shortcomings with the depreciation and amortisation regime. They merely added new provisions to the old, dealing with a handful of identified omissions rather than providing a sound general conceptual framework for amortisation of wasting expenditures. Plain English rules translating unsound and incoherent rules to a new format failed completely to address the underlying problems.
(ii) Capital Gains
The limitations of redrafting were even more evident in the case of the capital gains rules. The initial judicial distinction between income gains and capital gains, with only the latter subject to tax, was clearly incompatible with a modern income tax system. But rather than modify the definition of income to displace the judicial distinction between taxable income and non-taxable capital gains, the legislature adopted the judicial distinction as the basis for legislative response and embarked on several decades of ad hoc and piecemeal inclusion measures — each directed at a particular type of gain — to overcome the limitations of the judicial distinction.
The policy proved a failure, as the boundaries for each inclusion provision were easy to circumvent. In 1985 the legislature tried a different approach, relying on a broader inclusion measure. The legislative solution was built around a single provision aimed at one common type of capital gain, the gain realised on the sale of investment assets. The rule treated the difference between the cost of acquisition of an asset (inflation adjusted in some cases) and the proceeds realised on disposal of the asset as a capital gain. Complex deeming provisions were then added to bring the vast range of other types of capital gains within this rule. These supplementary measures deemed taxpayers in receipt of other types of ‘capital gains’ to have acquired property and then to have immediately disposed of the property. They also provided for deemed costs of acquisition and deemed proceeds of disposal in respect of the deemed acquisition and disposal of property. It surprised no one when the courts declined to apply the provisions as originally drafted, declaring they were virtually meaningless.
The 1997 plain English rewrite substituted a series of ‘capital events’ for the deemings on top of deemings aimed at different types of capital gains. As drafting proceeded, the drafters discovered they needed increasing numbers of events to catch all the different transactions and arrangements that might give rise to a capital gain in the judicial sense. By the time the redrafted capital gains measures were enacted into law, they contained 36 separate capital gains events, since expanded to 40 events. The measures were without doubt easy to read. Taken on their own, each of the events was comprehensible. When read together with the remainder of the legislation they created a complicated web of overlaps and distinctions. A gain might be assessable as ordinary income under the judicial tests, as statutory income under a specific inclusion provision, and again as a capital gain under one or more capital gain event. Reconciliation provisions were needed where more than one section applied to a gain in one tax year and timing provisions were needed where different assessment provisions would cause the same gain to be recognised in more than one tax year. The final result was certainly no better than its predecessor and may have been worse.
(iii) The Legacy of Plain English
The plain English simplification project showed that poor drafting had made the tax law difficult to understand. Abandoning archaic language and rigid drafting in favour of simpler language, clearer explanation and flexible presentation led to shorter sections (in some cases replacement measures were only half the size of the original sections) and law that was far easier to understand. It was hoped that better drafting would be one lasting legacy of the project but more recent versions of plain English draft raise serious doubts as to whether logical drafting will continue. Perhaps the most important contribution of the simplification project was to expose the fact that drafting complexity was not the cause of tax complexity. Bad drafting may have made the sections difficult to understand, but it could not obfuscate the underlying principles of the law if there were none. The new law appeared on its face to be easier to read compared with the original legislation, but, as one critic put it, without major policy changes the project could accomplish little more than ‘easier to read gibberish’.
The government never explicitly conceded the validity of the growing chorus of criticism for its redraft project, but by the time the second tranche of plain English legislation had been released it was obvious to everyone that clearer language was not bringing clearer tax law. In mid-1998, when only one-third of the law had been replaced with plain English text, the government quietly shelved the simplification project in the context of appointing a Review of Business Taxation. The redraft simplification project has never been formally revived. New legislation subsequently inserted in the 1997 Act has followed the general style of the ‘plain English’ draft but any rhetoric that suggests plain English drafting might yield simpler legislation has been abandoned. And, as noted in section 2C, above, so, too, has any commitment to genuine plain English with the move to defining terms for tax purposes to mean other than, or even opposite to, their ordinary English meaning.
Australia’s second large-scale proposal for tax simplification, the tax value proposal, derived from a Coalition Government’s tax reform election manifesto released in June 1998. A key element of the manifesto was a promise to establish a business-oriented review of business taxation in Australia. The Review of Business Taxation (known as the Ralph Review after its chair, businessman John Ralph) was appointed soon after the Coalition’s re-election later in 1998 and was asked to report in the first quarter of 1999. By early 1999 it was clear the reporting deadline could not be met and John Ralph sought an extension to mid-year. In return, he offered to have his review prepare draft legislation to implement any key recommendations of the project so they could be enacted more quickly if the government accepted the review’s recommendations.
The draft legislation released with the final report of the Ralph Review did not merely contain amendments to implement the review’s recommendations. Rather, it provided a blueprint for a complete redraft of the income tax Act, one that would replace all the conventional rules for measuring taxable income with a dramatically different formula. Ralph, speaking as chair of the business tax review, promised that the new law would be the key to tax simplification, a position subsequently echoed by many proponents of the new law in the Treasury and Australian Taxation Office.
The proposed new rules were based on what became known as the tax value method. Under the tax value rules, a taxpayer would have recognised annual gains and losses by recording all economic flows received and paid (in cash or in kind) and computing changes in the value of assets owned by a taxpayer at the end of the year. Thus, for example, rather than denying taxpayers an immediate deduction for the acquisition of a capital asset, as the current law now does, the proposed law would have allowed an upfront deduction matched by an offsetting inclusion of the increased asset value of the property acquired (from zero before the acquisition to its market value when acquired).
There is no doubt that the proposed rules would yield a much more accurate measurement of true economic income were they to be applied consistently to all changes in value of taxpayer’s assets. This is not what the Ralph Review proposed. Rather, it proposed to apply the existing depreciation rules to mandate presumptive changes in value of depreciable assets, the existing optional closing valuation rules to determine changes in value of trading stock, the existing amortisation rules for discounted debt instruments to measure gains and losses over time for debt assets and liabilities, and the existing realisation basis for virtually all other assets by deeming the closing value of the assets on hand to be equal to their cost. In other words, the tax value method would have incorporated all the existing rules for different types of assets and retained all the ambiguous and uncertain asset boundaries (trading stock vs. depreciable property vs. financial assests vs. other capital assets, etc.) of existing law. Stripped of the new foundation provisions, the emperor’s new suit looked much the same as the current one.
Equally significantly, the review proposed to retain all existing tax expenditures and concessions within the new structure if the tax value basis for calculating taxable income were adopted. It also proposed to incorporate into the tax value method approach a slew of new concessions promised by the government such as the subsequently adopted concessional regime for capital gains.
Whatever the potential of the tax value method for simplifying tax law, the benefits would have instantly dissipated once it incorporated all current concessions, in particular concessions such as the capital gains concession based on indeterminate judicial notions. A not insignificant part of the complexity in the existing tax law is attributable to the anti-avoidance provisions directed at schemes to exploit the concessional treatment of undefined capital gains and the same would surely have occurred with a tax value method tax law that incorporated a similar concession.
Given the loud and virtually unanimous opposition to the tax value method proposals from the profession and a cross-section of other interested persons including a number of influential tax academics, the lengthy demise of the tax value proposals was perhaps surprising. The explanation probably lies in the political sphere — the government of the day had made a range of promises to obtain opposition support in the Senate for its most significant tax concessions and it had to keep alive the illusion of future reforms for some period after its initial concessions were enacted. It allowed the tax value method proposal to pass gracefully, sending it to a quasi-independent ‘Board of Taxation’ for lengthy analysis while funding ongoing revisions of the tax value method drafts for almost two years. The tax value method’s official death was announced in mid-2002, following release of the Board’s final report on the proposal.
There are two clear lessons to be learned from the tax value method episode. The first is that simplification based on wholesale change of tax laws and tax principles is doomed to failure. Grand plans will inevitably be derailed by a combination of vested interests and taxpayers’ fears of transitional and implementation costs. The second is that incremental change, a far more realistic option, will only untangle complexity if it recognises and addresses the root causes of that complexity. The causes can never be eliminated — judges will continue to rely on inappropriate transposed categories to interpret tax law, tax advisers will continue to press the boundaries in the law, and legislatures will continue to use the tax Act as a vehicle for subsidies. In short, the causes of complexity will always be with us. But while the causes will never disappear, they can be tamed.
Where does an achievable path to simplification start? The review of the causes of complexity in section 2 of this paper may yield some practical steps that can be taken to achieve genuine simplification in a manner acceptable to the political masters of the legislation.
(i) Simple Solutions to Simple Problems: Addressing Structural Flaws
Many of the problems the tax value method purported to address had been long recognised and in terms of both their origin and nature were relatively simple. The tax value method solution to those problems was wholesale replacement of current tax concepts with a new system for measuring net gains. But comprehensive solutions to the key problems that yield much of the complexity in current law can be achieved in an incremental fashion by making relatively minor changes to both the income and deduction sides of the current net taxable income equation.
The process may be illustrated with the example of capital expenses. As noted earlier, there are several problems with the present Australian law relating to capital expenses. The first is that the distinction between current expenses (those consumed within the year of acquisition) and capital expenses (those yielding a longer lasting benefit) is based not on actual analysis of the benefit or asset acquired. Instead, it is based on transplanted judicial criteria that sometimes yield appropriate characterisations and often do not. In some cases, the judicial tests permit taxpayers to deduct currently capital expenses yielding long term intangible benefits. More significantly, they often characterise as capital expenses whose benefit expires almost immediately. These will never be recognised for tax purposes if they do not fall into one of the very limited group of deduction rules for ‘capital’ expenses with no lasting benefit.
The second problem with the current law is its inability to accommodate many outlays for long-term wasting benefits. The ad hoc and unprincipled depreciation rules in the current legislation add significant complexity to the tax law but fail to cover the field. Taxpayers often discover none of the depreciation provisions apply to their expenditures for wasting long term benefits and the outlays fall into a black hole for tax purposes.
The tax value method proposals sought to address both these problems by measuring cash flows and the value of benefits acquired with expenditures instead of characterising the expenditures. A capital expenditure that yields a long term benefit of quantifiable value would be recognised by means of a deduction for the outlay offset by the inclusion of the value of the benefit acquired. Taxable income would be reduced each year as the benefit declined in value.
There are two crucial differences between the tax value method proposals and current law. The first is the test used to distinguish capital and current expenditures. The tax value method proposals used an objective test to identify a capital expense — did the expense result in the acquisition of a benefit with recognisable value lasting beyond the end of the tax year? The judicial tests used in the current tax law are based on a range of factors, largely derived from trust law precedents originally used to identify the class of beneficiary that should bear costs incurred by a trust in respect of assets for the remainder beneficiaries. From a policy perspective, therefore, the initial problem with current law is a definitional one — the definition of capital expenses includes some outgoings that should be deductible immediately and excludes some outgoings that should be recognised in a future year or years, as the assets they yield waste or when there is a disposal of those assets.
Rather than rewrite the entire Income Tax Assessment Law to correct indirectly the definition of a capital expense, a new statutory definition can be substituted that makes it clear a capital expense is one which yields an asset or benefit with a quantifiable life extending past the end of the income year. Such a definition would mean that expenses whose benefit expired almost immediately would be deductible when the outgoings were incurred. At the same time, it would prevent the misclassification as revenue expenses outgoings that yield long-term benefits.
The second key difference between the tax value method and current law is that the underlying accounting formula in the former provided recognition for all expenses incurred to acquire wasting assets while the limited depreciation regime in the latter combines with the structural rigidity of the capital gains rules to banish many types of capital expenses to legislative lacunae, never to be recognised for tax purposes.
If an appropriate statutory definition of capital expenses were adopted to substitute for the current judicial definition, the current capital allowance rules could easily be supplemented with a final catch-all rule that allows taxpayers to amortise the cost of all wasting assets or benefits not falling into other capital allowance schedules over their effective lives.
A simple start such as signalling to the courts an appropriate test to distinguish capital and revenue expenses and correcting a gap in the law will open the door to many opportunities for further (and incremental) simplification. Once a rational definition of capital expenses is adopted, and a simple supplementary rule to guarantee comprehensive recognition of those outgoings and to eliminate black hole expenses is inserted in the legislation, the complex infrastructure of provisions on top of provisions aimed at overcoming the problems caused by the current judicial distinction between revenue and capital expenses can gradually be dismantled.
Relatively simple changes can establish a sound structural base for long term simplification. The same process can be applied to almost all the structural problems that give rise to complexity in the income tax law.
(ii) Simple Solutions to Simple Problems: Refining Technical Features of Tax Law with Purposive Definitions
The structural problem arising from the current definitions of capital and revenue expenses derives largely from inappropriate judicial tests used to distinguish the two types of expenses, tests transplanted from another area of law (a separate but related problem is the inadequate amortisation regime in the legislation). Transplanted definitions also cause serious problems with the operation of technical measures in the income tax law. This type of measure is purely a creature of statute and there is no reason why the legislation establishing technical rules cannot define central terms in a manner consistent with the objective of the rules.
The process can be illustrated with the term ‘employee’ which, as noted earlier, has been defined by the courts for tax purposes using precedents drawn from other areas of law. Tax law distinguishes between employees and independent contractors to determine whether the person should be subject to withholding tax based on gross income (employees) or be responsible for lodging provisional returns and tax payments based on estimated net income (independent contractors).
Currently, the judicial definition of employee for tax purposes derives from industrial law and tort law, particularly the latter. A key test in the current tort law definition is the extent to which an ‘employer’ controls the day-to-day activities of an employee. This test is clearly relevant to whether vicarious liability should be imposed on an impugned employer for negligence of an alleged employee — if the labourer worked under the direction or supervision of a principal, the principal should be able to minimise the risk of negligence and should be held accountable when there is negligence. If the labourer works independently, responsible for an output without direct supervision over how that output is generated, the principal cannot effectively minimise the risk of negligence by the labourer. If vicarious liability is imposed on employers for the actions of employees, it is logical to treat this latter labourer as an independent contractor and not an employee for tort law purposes.
Any overlap between the tests used in tort law to identify employees and the reason employees are categorised as a separate type of taxpayer for income tax purposes is at best the result of happy coincidence. For tax purposes, the test for identifying an employee should be based on the level of expenses incurred by the person and the possibility of those being passed on to the principal if tax is based on gross income. If a person incurs high costs in deriving income, a withholding tax on gross income will be unfair until the final reconciliation at the end of the year so designation as an independent contract would be appropriate, while if a person is able to require the ‘employer’ to bear the costs associated with his or her labour, designation as an employee would be most sensible from a tax administration perspective. Replacement of inappropriate tort law tests based on level of supervision and independence of a person with a statutory definition for tax purposes that sets out the objective of the definition and provides more appropriate tests could ameliorate significantly the problems currently encountered as a result of using the transplanted definition.
Simplifying tax law in terms of definitions can thus take dual tracks. First, where terms currently used are subject to continuing pressure (as evidenced, for example, by continuing litigation), purposive statutory definitions can be employed, at the same time that resulting redundant specific anti-avoidance rules are cleared out. Second, when new tax rules are adopted, they can be accompanied by purposive definitions. These can be combined with a policy of setting out the purpose of new measures so it made obvious that it would not be appropriate to adopt transplanted definitions intended for different purposes in other areas of law.
(iii) Taming Tax Expenditures
It was noted earlier that the uncertain borders delineating tax concessions usually result in unintended exploitation by taxpayers not engaged in activities or making investments of the sort the concession was intended to promote. Because the purpose of the concession is rarely spelled out in law, administrators and courts have no guidance as to what should or should not qualify for concessional treatment. Most often, initial concession measures are followed by a range of increasingly complex anti-avoidance rules intended to better focus the concession.
Rather than attack the problem after the fact, tax expenditures can be designed initially to operate effectively and simply without ambiguous borders if concessional regimes are defined in terms of their objectives, with a clear indication that eligibility for concessional treatment is to be based on underlying economic substance rather than legal form. If concessions were defined in terms of their purpose, courts would be directed to look at economic outcomes and not legal forms and the task of determining whether any particular arrangement, whatever its legal form, satisfies the objective of the concession would be simplified.
Some attempt has been made in the post-business tax review era to pay lip-service to the idea of including purposive guides in tax concessions. For example, the small business concessions introduced in 2001 have as a stated goal, among other things, the lowering of taxes for small businesses. This is the first time that a tax expenditure explicitly concedes that a primary purpose of the concession is to reduce taxes for the constituency at which the concession was aimed. However, the legislation gives no hint as to why taxes are to be lowered for this particular group instead of others. Rather than identify the intended beneficiaries in terms of an object to the concession, the legislation attempts to confine its application through the use of a complex qualification threshold based on average turnover and value of depreciable assets owned by the business (qualifying businesses must be below both thresholds). These tests are augmented by tracing rules and grouping rules intended to prevent large businesses from reorganising parts into smaller and apparently separate qualifying small businesses. The result is several pages of complex legislation yielding substantial compliance burdens legitimate for many small businesses seeking the benefit of the concession and a technical roadmap for non-qualifying businesses wishing to rearrange in order to qualify for them.
Plausible grounds for the small business concession can be posited. For example, because of economies of scale when dealing with financial institutions and the costs of complying with securities laws when raising public funds, small business often faces a significantly higher cost of capital than large business. Where funding is crucial to business operations and expansion, as is the case in capital intensive activities such as manufacturing, primary production, and mineral extraction, small business is prejudiced compared to its larger business counterparts. Lower taxes would provide small business relying on retained earnings as a source of capital with less expensive funds. However, if the purpose of the lower tax rate were to reduce the cost of capital for deserving small business, it is unlikely that the concession would be designed to apply equally to, say, small manufacturers with significant capital needs and small independent consultants with little or no capital needs.
The current test for eligible small business, based on turnover below a threshold and depreciable asset holdings below a threshold, is unlikely to be an effective indicator of eligibility if the concession is intended to achieve rational economic objectives such as lowering the cost of capital for businesses at a relative disadvantage in obtaining lower cost capital. Alternative tests based on rational criteria such as operational (as opposed to passive) asset holdings are more likely to complement genuine economic goals, if there are any that explain the concession.
The clearest roadmap for tax administrators and adjudicators would be one that sets out the purpose of the concession and explicitly identifies the type of taxpayer for which it is intended, with the identification being framed in terms of relevant economic criteria such as indicia of business size that are directly tied in the legislation to the identified object of the concession. Thus, for example, if turnover were to be used as one test for qualifying businesses, it should be made clear that low-margin businesses with high turnover may still qualify while high margin businesses with lower turnovers may not.
To be sure, reforms to existing concessions that involved identification of the purpose of the concessions would alter the tax base. Consider, for example, the capital gains concession introduced in 1999 (a 50 per cent exemption from tax for realised capital gains). The rationale for the concession was never articulated and apart from its obvious effect — to reduce the tax burden for highest income individuals who are able to realise much if not most of their income as capital gains — it is difficult to posit a convincing purpose for the concession. However, subsidiary sources including political speeches suggest a rationale may be to promote capital investment in industrial, commercial or primary production enterprises. If this explanation were adopted for the moment and the concession redrafted as one designed to achieve this end, some types of gains currently said to have a capital character would lose their concessional treatment. Losers might include gains on the sale of personal art collections and on the sale of personal-use real estate such as summer homes.
Of course, the government could easily extend the capital gains concession to these and other gains if it concluded investments in assets yielding these gains should be encouraged. But setting out the concession in these terms — drafting a purpose for the concession and identifying what type of income, in the generic sense, should qualify for the concession — would put a stop to most schemes and arrangements to recast judicial concept income as concessionally taxed capital gains. By reclaiming from the judiciary the power to determine which gains should be fully taxed and which gains may be concessionally taxed, the legislature will provide a path to simpler tax law, tax administration, and tax compliance. The importance of this shift in political terms, too, cannot be understated. Traditionally, commentators have defined legitimate taxation in terms of the government’s onus to specify when income is included in the tax base. Similar standards should be imposed on concessions providing relief from full taxation to establish the legitimacy of the concessions and to reinforce the legitimacy of the rules that impose full taxation on comparable economic income derived by those unable to access concessions. In the absence of specificity on the concession side, there is a serious danger that the tax law will be perceived as little more than a collection of unfair and illegitimate ad hoc inclusion and concession measures.
The causes of complexity in the Australian income tax are varied. Taxpayers and tax authorities have long recognised the law is in need of simplification. During the past decade there have been two ineffective or unsuccessful attempts at simplification of the tax law, first by redrafting the law in plain English without addressing structural issues and second by the proposed wholesale replacement of the legislation and its current concepts with a new foundation that incorporated most of the causes of complexity in the current law. In the meantime, the law continues to grow in size and complexity. It is time to consider new paths to simplification, learning from the causes of complexity and past attempts at simplification. Simple amendments to address structural flaws, purposive definitions and better-designed tax expenditures are useful starting points for such an exercise.
[*] © R. Krever, 2003.
Professor of Law, Deakin University, Melbourne. Many of the ideas in this article derive from a paper presented by the author at a taxation symposium organised by the University of Potsdam, Germany in 1999, although its resemblance to that paper has faded significantly as a result of more than three years of revision. Richard Haigh, Michael Kobetsky and Justice Hill provided much welcomed criticism of that original manuscript, available in Hans-Georg Petersen & Patrick Gallagher (eds), Tax and Transfer Reform in Australia and Germany (2000) chapter 5, ‘Simplicity and Complexity in Australian Income Tax’.
 The leading (and perhaps definitive) Australian analysis of simplification issues contains an excellent summary of simplification concepts and the causes of complexity — see Graeme Cooper, ‘Themes and Issues in Tax Simplification’ (1993) 10 Australian Tax Forum 417. A helpful legal-economic review of the issues is found in Binh Tran-Nam, ‘Tax Reform, Tax Simplification: Some Conceptual Issues and Preliminary Assessment’  SydLawRw 20; (1999) 21 Syd LR 500.
 See John Freebairn, ‘Microeconomic Reform and Tax Simplification’ (1993) 10 Australian Tax Forum 461 at 464 and the sources cited therein.
 See, for example, Jeff Pope, Richard Fayle & Mike Duncanson, The Compliance Costs of Personal Income Taxation in Australia, 1986/87 (Sydney: Australian Tax Research Foundation, 1990); Jeff Pope, Richard Fayle & Dongling Chen, The Compliance Costs of Public Companies’ Income Taxation in Australia, 1986/87 (Sydney: Australian Tax Research Foundation, 1991); Jeff Pope, Richard Fayle & Dongling Chen, The Compliance Costs of Public Companies’ Income Taxation in Australia (Sydney: Australian Tax Research Foundation, 1994); Chris Evans, Katherine Ritchie, Binh Tran-Nam & Mike Walpole, Costs of Taxpayer Compliance (Canberra: AGPS, 1996); Chris Evans, Katherine Ritchie, Binh Tran-Nam & Mike Walpole, A Report into the Incremental Costs of Taxpayer Compliance (Canberra: AGPS, 1997); Chris Evans, Katherine Ritchie, Binh Tran-Nam & Mike Walpole, A Report into Taxpayer Costs of Compliance (Canberra: AGPS, 1997); Chris Evans, Katherine Ritchie, Binh Tran-Nam & Mike Walpole, ‘Taxation compliance cost: some recent empirical work and international comparisons’ in Chris Evans & Abe Greenbaum (eds), Tax Administration: Facing the Challenges of the Future (1998) at 177.
 See Ian Wallschutzky, ‘TLIP: Stage 1 — Benchmarking’ (1995) 12 Australian Tax Forum 115.
 While it is far larger than the legislation of most of Australia’s main trading partners, it admittedly exceeds the US law in length only because much of the detail of US law appears in regulations rather than the legislation.
 Preliminary results from a study carried out by Mike Walpole and others indicate that readers found the post-1997 drafting style easier to understand (working papers available from email@example.com). The first drafts in the new style dealt with relatively simple areas of law. A far more complex style has been used for more recent amendments.
 Treasury, Reform of the Australian Tax System: Draft White Paper (Canberra: AGPS, 1985) at 1.
 Since 1997 there has been two separate income tax Acts in place. However, since 1999 there has been virtually no new legislation related to the migration of measures from the old law to the new one; the rate of growth continues to expand nevertheless.
 Four colonies levied income taxes before federation. See further Ian van den Driesen & Richard Fayle, ‘History of Income Tax in Australia’ in Rick Krever (ed), Australian Taxation: Principles and Practice (1985) at 28–29 and Peter Harris, Metamorphosis of the Australasian Income Tax: 1866 to 1922 (Canberra: Australian Tax Research Foundation, 2002).
 A particularly vocal critic was Yuri Grbich. See, for example, Yuri Grbich, ‘Problems of Tax Avoidance in Australia’ in John Head (ed), Taxation Issues of the 1980s (Sydney: Australian Tax Research Foundation, 1983) at 413; Yuri Grbich, ‘Section 260 Re-examined: Posing Critical Questions about Tax Avoidance’  UNSWLawJl 3; (1976) 1 UNSWLJ 211; Yuri Grbich, ‘Anti-avoidance Discretions: The Continuing Battle to Control Tax Avoidance’  UNSWLawJl 11; (1981) 4 UNSWLJ 17.
 The term is attributable to Neil Brooks in his outstanding study of the role of judicial decision-making in creating tax law complexity, ‘The Role of the Judges’ in Graeme Cooper, Tax Avoidance and the Rule of Law (1997) at 122.
 For a full analysis of the transplanted trust law concepts, see Ross Parsons, ‘Income Taxation: An Institution in Decay?’ (1986) 3 Australian Tax Forum 233.
 See, for example, Norman v FCT  HCA 21; (1963) 109 CLR 9, where the High Court concluded it was not possible to shift liability for tax on dividends by assigning the right to dividends to a spouse, Shepherd v FCT  HCA 70; (1965) 113 CLR 385 where the same court said it was possible to shift tax liability on royalties by assigning that right, and FCT v Everett  HCA 6; (1980) 143 CLR 440 where the High Court said it was possible to assign income from a professional partnership to a spouse.
 See, for example, World Book (Australia) Pty Ltd v FCT (1992) 23 ATR 412, placing considerable reliance on Stevens v Brodribb Sawmilling Co Pty Ltd (1986) 160 CLR 16.
 See, for example, FCT v Murry  HCA 42; (1998) 39 ATR 129, drawing upon Geraghty v Minter (1979) 142 CLR 177.
 The term derives from a description of Jordan CJ in Scott v Commissioner of Taxation (NSW)  NSWStRp 9; (1935) 35 SR (NSW) 215; 3 ATD 142, interpreting the meaning of ‘income’ for the purpose of a State income tax law. His phrase achieved universal acceptance by the courts and in 1997 a variation was inserted in the income tax law, which now refers to judicial concept income as ‘ordinary income’ (see Income Tax Assessment Act 1997 s6-5).
 One of the leading judicial precedents on the role of intent in characterising income is FCT v Whitfords Beach Pty Ltd  HCA 8; (1982) 150 CLR 355. Not long before it sold some property it had owned for years, the taxpayer in that case changed its intent and manner of dealing with the property. The High Court said the property should be valued at the time the taxpayer’s intention changed, with the gain measured from original cost to the change of intent value being a non-taxable capital gain and the gain from value at the time of intent change to sale being an income gain.
 Neil Brooks argues that the courts should have independently examined the reason for the use of the term and developed a meaning appropriate for tax law. See Brooks, above n11.
 [Former] Income Tax Assessment Act 1997 s118-250.
 Goodwill now qualifies for the small business ‘active asset’ concessions in Division 152 Income Tax Assessment Act 1997 (see s152-40(b)).
 Public finance scholars have long argued, based on the work of Simons and Haig, that the decision to realise or not realise appreciated gains or losses was a matter of taxpayer portfolio choice and for equity and efficiency reasons, unrealised appreciation and depreciation should be recognised for tax purposes in theory, though they generally concede a full accrual regime applying to all unrealised gains and losses might be impossible to achieve in practice. See Henry Simons, Personal Income Taxation: the Definition of Income as a Problem of Fiscal Policy (1938) at 49, and also Robert Haig, ‘The Concept of Income: Economic and Legal Aspects’ in Robert Haig (ed), The Federal Income Tax (1921).
 For example, the Australian judicial characterisation of fines as non-deductible quasi-personal expenses (even for a corporation) is based on two first instance English cases, IRC v Warnes & Co  2 KB 444 and IRC v von Glehn & Co Ltd  2 KB 553. See, for example, Madad Pty Ltd v FCT  FCA 287; (1984) 15 ATR 1118 and Mayne Nickless Ltd v FCT (1984) 84 ATC 4458.
 In the 1980s, the UK judiciary shifted ground significantly and developed general anti-avoidance doctrines known as the fiscal nullity doctrines (see WT Ramsay Ltd v IRC  UKHL 1;  AC 300). Australian courts said the new UK approach had no application in Australia, a position finally confirmed by the High Court in FCT v John (1987) 19 ATR 150; 82 ATC 4713. There have, however, been instances in recent times outside the anti-avoidance field where Australian doctrines are diverging from UK precedents. See, for example, Steele v DCT  HCA 7; (1999) 197 CLR 459, 41 ATR 125 where the approach taken by the Privy Council in Wharf Properties Ltd v Commissioners of Inland Revenue of Hong Kong (1997) 97 ATC 4225 was distinguished.
 See, for example, the joint judgment of Bowen CJ, Lockhart & Gummow JJ in All States Frozen Foods Pty Ltd v FCT (1990) 20 ATR 1874.
 See Higgs v Olivier  1 Ch 899 in which an English actor was paid in part for his appearance in a film and in part for his agreement not to produce films for other studios that might compete with the film.
 Kitto J in Dickenson v FCT  HCA 62; (1958) 98 CLR 460; 7 AITR 257 at 280.
 Definition of ‘eligible termination payment’ in Income Tax Assessment Act 1936 s27A(1).
 Definition of ‘fringe benefit’ in Fringe Benefit Tax Assessment Act 1997 s136(1).
 Income Tax Assessment Act 1997 s104-35.
 The phenomenon is well illustrated by the adoption of Income Tax Assessment Act 1936 s26(a), enacted in response to the UK House of Lords decision in Jones v Leeming  AC 415;  All ER 584. When the Australian courts read down the measure in a manner that threatened to constrict the tax base significantly, the short and simple measure was expanded (and renumbered as s25A) to include several pages of subsections, each aiming at one or more judicial precedents that had narrowed the scope of the original measure.
 Inland Revenue Commissioners v Duke of Westminster  AC 1. While neither the first nor the most significant avoidance case relying in part on strict literalism for the avoidance to be effective, the Duke of Westminster rose quickly to prominence as the authority for literalism and the statement of Lord Tomlin (at 19): ‘Every man is entitled if he can to order his affairs so as that the tax attaching under the appropriate Acts is less than it otherwise would be’ remains one of the most cited passages in Australia from UK tax cases.
 See further, Robert Allerdice, ‘The Swinging Pendulum: Judicial Trends in the Interpretation of Revenue Statutes’ (1996) 9 UNSWLJ 162; Douglas Brown, ‘The Canons of Construction of Taxation and Revenue Legislation’ (1976) 5 AT Rev 81; Michael D’Ascenzo, ‘Along the Road to Damascus: a Framework for Interpreting the Tax Law’ (2000) 3 J of Aust Taxation 384. The debate is not confined to Australia — see, for example, Michael Livingston, ‘Practical Reason, “Purposivism,” and the Interpretation of Tax Statutes’ (1996) 51 Tax Law Rev 677; Lawrence Zelenek, ‘Thinking About Nonliteral Interpretations of The Internal Revenue Code’ (1986) 64 North Carolina Law Rev 623; Brian Arnold, ‘Reflections on the Relationship Between Statutory Interpretation and Tax Avoidance’ (2001) 49 Canadian Tax J 1; John Avery Jones, ‘Tax Law: Rules or Principles?’ (1997) 17 Fiscal Studies 63; Natalie Lee, ‘A Purposive Approach to the Interpretation of Tax Statutes?’ (1999) 20 Statute Law Rev 124; Robin Burgess, ‘Form Without Substance? A Comment on Tax Avoidance and its Influence in the Interpretation of Tax Statutes’ (1982) Statute Law Rev 87.
 See, for example, Geoffrey Lehmann, ‘The Income Tax Judgements of Sir Garfield Barwick: A Study in the Failure in the New Legalism’  MonashULawRw 1; (1983) 9 Mon LR 115.
 For a very contrary view, see Graham Hill, ‘Barwick CJ: “The Taxpayer’s Friend?”’ (1997) 1 The Tax Specialist 9.
 A representative example may be found in FCT v Westraders Pty Ltd  HCA 24; (1980) 144 CLR 55. The Chief Justice, with whom all judges but one concurred, began his judgment in favour of the taxpayer with the observation that, ‘the facts of this case disclose an ingenious use of the provisions of sections 36 and 36A of the Income Tax Assessment Act 1936’.
 The controlled foreign corporation (CFC) rules using 130 pages to cover what should be a fairly simple concept provides a fine example of such legislation.
 That is, a blended payment loan such as a standard home mortgage loan in which principal is repaid along with interest over the life of a loan.
 ANZ Savings Bank Ltd v FCT (1992) 24 ATR 201; 92 ATC 4630 (Jenkinson J).
 ANZ Savings Bank v FCT  FCA 282; (1993) 42 FCR 535; 25 ATR 369; 93 ATC 4370 (Hill J with Heerey J agreeing & Davies J dissenting).
 Subsequent to the ANZ Savings Bank case the legislature finally acted, pulling commercial annuities of the sort encountered in that case out of the annuity rules and treating them instead as ordinary blended payment loans for tax purposes — these annuities are defined as ‘ineligible’ annuities in Income Tax Assessment Act 1936 s159GP(1) and then taken out of the annuity treatment by Income Tax Assessment Act 1936 s27H(4).
 FCT v Rowe  HCA 16; (1997) 187 CLR 266.
 For some judges’ views on interpretation issues, see Sir Ivor L M Richardson, ‘Appellate Court Responsibilities and Tax Avoidance’ (1985) 2 Australian Tax Forum 3; Graham Hill, ‘The Judiciary and its Role in the Tax Reform Process’ (1999) 2 Journal of Aust Taxation 66. For a UK perspective often considered in Australia see Sir Peter Millett, ‘Artificial Tax Avoidance — The English and American Approach’ (1988) 5 Australian Tax Forum 1.
 Interestingly, s15AA of the Acts Interpretation Act 1901 (Cth), inserted into the Act at the height of the 1970s/early 1980s tax avoidance era, has had virtually no effect on judicial doctrines. The section reads, ‘In the interpretation of a provision of an Act, a construction that would promote the purpose or object underlying the Act (whether that purpose or object is expressly stated in the Act or not) shall be preferred to a construction that would not promote that purpose or object’.
 One commentator placed the cost at $10 billion — see Yuri Grbich, ‘Problems of Tax Avoidance in Australia’ in John Head (ed), Taxation Issues of the 1980s (Sydney: Australian Tax Research Foundation, 1983) at 413. Selected statistics illustrating the magnitude of the problem may be found in Richard Krever, ‘Tax Reform in Australia: Base Broadening Down Under’  34 Can Tax J 346 at 352.
 Mark Leibler, ‘Should the Law Institute of Victoria Take a Public Stand Against Artificial and Contrived Tax Avoidance Schemes — A Debate’ (1980) 54 Law Inst J at 562–564.
 It was assumed that the bias to retention in public companies would be offset somewhat by demands for distributions by investors, particularly institutional investors.
 At one end of the scale there was a distinction — one type of public company was a company listed on a public stock exchange. However, the definition of public companies extended to a range of privately held entities as well.
 The leading case, Slutzkin v FCT  HCA 9; (1977) 7 ATR 166 became a catchword for various tax avoidance schemes.
 FCT v Peabody (1994) 181 CLR 359.
 FCT v Spotless Services Ltd  HCA 34; (1996) 186 CLR 404.
 Truesdale v FCT  HCA 27; (1970) 120 CLR 353.
 ANZ Savings Bank Ltd v FCT (1992) 24 ATR 201. See discussion, above n38.
 A leading international expert reviewing the Australian experience at an international conference in 1990 described the quality of Australian tax drafting as ‘abysmal’ — see Ferrers, ‘Towards Making Our Tax Law Understandable’ (1990) 2 The CCH Journal of Australian Taxation 98 at 98, quoting Canadian law professor Brian Arnold. A classic example later cited by a tax simplification team as an example of the problem is former paragraph 82KTJ(4)(b) of the Income Tax Assessment Act 1936: ‘a car is a predecessor of a second car if a third car is predecessor of a second car and the first-mentioned car is a predecessor of the third car (including a case where the first-mentioned car is a predecessor of the third car by another application or applications of this paragraph).’ Various courts have been direct in their criticism of the drafting — Hill J of the Federal Court noted one section on which he was asked to rule had been drafted ‘with such obscurity that even those used to interpreting the utterances of the Delphic oracle might falter in seeking to elicit a sensible meaning from its terms’: FCT v Cooling (1990) 90 ATC 4472 at 4488. The unintelligibilty of the law and the potential for improvement through redrafting was set out convincingly in Christopher Balmford, ‘Redrafting the Income Tax Assessment Act 1936 in Clear Language’ (1991) 2 Rev LJ 1.
 Consolidated Press Holdings v FCT (1998) 98 ATC 5009 at 5018.
 The project is described in detail in section 3A, below.
 Income Tax Assessment Act 1997 s115-1.
 The background to this phenomenon is discussed below in section 3B(i).
 Capital Gains Taxes, Treasury Taxation Paper No 10, reprint of a Treasury submission to the Taxation Review Committee (Asprey Committee) (Canberra: AGPS, 1974) at 4. Treasury advised that ‘dispensing with the distinction between income and capital profits’ would further the goal of income tax simplicity.
 Stanley S Surrey, Pathways to Tax Reform: The Concept of Tax Expenditures (1973).
 Treasury, Tax Expenditures Statement 1997–98 (Canberra: AGPS) at 13, 58; Australia, Budget Paper No 1 1999–00 (Canberra: AGPS) at 6-23.
 Initially, three concessions were adopted: indexation of the cost base, averaging of the gains, and negative gearing whereby interest expenses incurred to derive gains are immediately deductible while recognition of the resulting gain is deferred until realisation. The first two benefits have recently been replaced with a flat 50 per cent exemption for gains realised by individuals on assets held for a year or more prior to disposal.
 Division 152-B, Income Tax Assessment Act 1997.
 Division 152-C, Income Tax Assessment Act 1997.
 Division 152-D, Income Tax Assessment Act 1997.
 Division 152-E, Income Tax Assessment Act 1997.
 Division 328, Income Tax Assessment Act 1997.
 Section 328-5, ITAA 1997.
 Harry Gibbs, ‘The Need for Taxation Reform’, Taxation Institute of Australia 11th National Convention (1993) 15 at 18; see also Harry Gibbs, ‘The Tax System: Seriously Wrong in Principle’ (1993) (July) Taxation in Australia 31 at 35.
 See ‘Tax Simplification’, statement by the Treasurer in (1990) 24 Taxation in Australia 602.
 Joint Committee of Public Accounts, A Report on the Administration of Tax Law and the Operations of the Australian Tax Office Report No 326 (Canberra: 1993).
 See ‘The Tax Law Improvement Project: Mission Impossible?’ (1994) 28 Taxation in Australia 484.
 The objectives of the project were outlined by the project’s director and its chief drafter in Brian Nolan & Tom Reid, ‘Re-writing the Tax Act’ (1994) 22 Fed LR 448. A brief summary of the early stages of the project may be found in Ian Wallschutzky, ‘The Tax Law Improvement Project: Australia’s Attempt to Simplify its Tax Laws’ (1995) 1 Asia-Pacific Tax Bulletin 119; A Towler, ‘Tax Law Improvement — An Overview’ (1994) 6 (5, Oct/Nov) The CCH Journal of Australian Taxation 14; A Towler, ‘Tax Law Improvement — Building the New Income Tax Law’ (1995) 7 (2, Apr/May) The CCH Journal of Australian Taxation 46.
 See, for example, Peter Cowdroy, ‘Dross into Gloss?’ (1995) 30 Taxation in Australia 187; Anthony Smyth, ‘The New Income Tax Assessment Act: After the Hype, the Old Complexities Remain’ (1995) (Jun/Jul) The CCH Journal of Australian Taxation 8. For a spirited defence of TLIP, see Brian Nolan, ‘I come to bury the 1936 Act’ (1997) 31 Taxation in Australia 570.
 See Mark Burton & Michael Dirkis, ‘The Income Tax Simplification Experience to Date’ (1996) 50 Bulletin for International Fiscal Documentation 67.
 The Income Tax Assessment Act 1936 became an alpha-numeric maze with more letters added each time a section was inserted between existing sections. Thus, ss159GZZZBA to 159GZZZBI were found between s159GZZZA and s159GZZZB, and so on through the maze of new provisions as whole divisions and new parts of the law were inserted between s159 and s160.
 Studies on the improved readability of the new law are reviewed in the somewhat misnamed Robin Woellner, Cynthia Coleman, Margaret McKercher, Michael Walpole & Julie Flynn, ‘Once more into the breach ... a study of comparative compliance costs under the 1936 and 1997 Acts: progress report’ in Chris Evans & Abraham Greenbaum, Tax Administration: Facing the Challenges of the Future (1998) at 195.
 M Evans, ‘Simplification’s Simple, Isn’t It?’, Taxation Institute of Australia South Australian State Convention (1997) 404 at 412.
 In one celebrated case, the owners of a building used to dye textiles characterised the entire building as equipment in a successful attempt to depreciate the entire structure: the floor was presented as dying pits, the walls as retaining devices to prevent dispersal of potentially toxic materials, and the ceiling as a ventilation system. See Wangarratta Woollen Mills Ltd v FCT  HCA 39; (1969) 119 CLR 1; 1 ATR 329.
 As noted below, for buildings alone there were multiple regimes for the mining industry, timber industry, hotel and motel industry, and more.
 Former Division 41, Income Tax Assessment Act 1997.
 The Review of Business Taxation or ‘Ralph Review’ — see section 3B below.
 For example, Income Tax Assessment Act 1936 s26AB included premiums payable in respect of a lease in assessable income while s26BB included small shallow premiums payable in respect of a debt in assessable income. No general provisions applied to large premiums, though they might constitute judicial concept income in some cases. If the debt was redeemed within a year of issue, the gain might have been assessable under Income Tax Assessment Act 1936 s26AAA. Alternatively, if the acquisition and redemption was part of a profit-making scheme or if the debt had been acquired with the purpose of resale at a profit, the gain might have been assessable under Income Tax Assessment Act 1936 s25A.
 Mason CJ described one section as ‘obscure, if not bewildering’ in Hepples v FCT 91 ATC 4808 at 4810; Spender J described the capital gains provisions as ‘nightmare provisions’ in Cooling v FCT 89 ATC 4731 at 4741. The confusion over the measures is quite evident in the High Court’s decision in Hepples v FCT 91 ATC 4808 where the judges agreed the provisions they were examining applied to some asset involved in the case, but couldn’t agree which asset the legislation was supposed to cover.
 Richard Vann, ‘Improving Tax Law Improvement: An International Perspective’ (1995) 12 Australian Tax Forum 193 at 218.
 Alice McCleary, ‘The Tax Law Improvement Project Debate: Aspects of the Law’ South Australian State Convention, Taxation Institute of Australia (1996) 84 at 88.
 See below, text at section 3B.
 Ralph Review of Business Taxation, A Tax System Redesigned (Canberra: AGPS, 1999).
 The Board of Taxation, Evaluation of the Tax Value Method: A Report to the Treasurer and Minister for Revenue and Assistant Treasurer (Canberra: Info Access, 2002).
 See above, section 3B(i).
 A leading example is Broken Hill Theatres v FCT  HCA 75; (1952) 85 CLR 423 in which a theatre owner incurred legal fees opposing a licence application by a potential competitor. While there was no known long term benefit from the expenditure (other persons remained free to apply for licences) the High Court characterised the expenditure as a capital outlay because it related to the taxpayer’s business ‘structure’ rather than its business ‘process’. Because the true value benefit may expire immediately and no asset is acquired, the expense can neither be depreciated nor recognised as a capital loss. Expenditures of this sort are commonly treated as currently deductible expenses in other jurisdictions.
 This example is drawn from Brooks, above n11.
 Section 328-50(1) Income Tax Assessment Act 1997 states: ‘The main object of this Division is to offer eligible small businesses the choice of a new platform to deal with their tax. The platform is designed to benefit those businesses in one or more of these ways: reducing their tax’.
 A recent enunciation of this view is that of Hill J: ‘It is, in my view, important in a democracy that the government be required to legislate with precision if it is to impose a liability upon its subjects’ in Justice Graham Hill, ‘A Judicial Perspective on Tax Law Reform’ (1998) 78 ALJ 685 at 689.