Treasury Economic Paper Number 12 – Some Economic Implications of Takeovers

PDF icon PDF 285KB

Treasury Economic Paper Number 12

SOME ECONOMIC IMPLICATIONS OF TAKEOVERS

Australian Government Publishing Service
Canberra, 1986
© Commonwealth of Australia 1986
ISBN 0 644 05048 9
Printed by C. J. THOMPSON, Commonwealth Government Printer, Canberra

FOREWORD

This paper, the twelfth in the series of Treasury Economic Papers, has been prepared at the request of the Treasurer in response to the recent public interest in, and discussion of, takeovers in Australia. The paper aims to contribute to this discussion by outlining some general and regulatory issues relating to takeovers and drawing out some of the principal economic and taxation implications.

The views expressed herein are those of the Department of the Treasury and are not necessarily those of the Government.

iii

CONTENTS

   

Page

     

FOREWORD

 

iii

     

1.

Introduction

1

     

2.

General Issues

1

 

2.1 Takeovers and the role of equity markets

1

 

2.2 The level of takeover activity

2

 

2.3 Objectives and effects of takeovers

4

     

3.

Some Implications

7

 

3.1 Effects on competition

7

 

3.2 Taxation aspects

8

 

Gearing and tax revenue

10

 

Imputation

11

 

Capital gains tax

12

 

Group loss provisions

12

 

Redeemable preference shares

13

 

Policy options

13

 

Indexation

13

 

Changes in interest receipts taxation

14

 

Thin capitalization rules

14

 

Quarantining of interest payments

16

 

Conclusion

19

 

3.3 Gearing

19

 

3.4 Diversion of funds from productive investment

20

 

3.5 External account issues

20

     

4.

Some Regulatory Issues

22

     
 

4.1 The role of corporate raiders and investment funds

22

 

4.2 Defensive measures by target firms

23

 

4.3 Partial bids

24

     

5.

Conclusion

26

     

Attachment A:

The cost to revenue of takeover activity

27

Attachment B:

Some data on gearing

34

Attachment C:

The section 46 dividend rebate

38

     

Appendix C1:

Sections 46 and 50: Legislation

41

Appendix C2:

Section 46: Developments in case law and legislation

43

Appendix C3:

Section 46 : Issues of principle

47

     

Attachment D:

Overseas tax experience

56

iv

1. INTRODUCTION

This paper reviews some general economic issues associated with company takeovers in the light of concerns that have been expresser, particularly about possible taxation and foreign exchange implications.

2. GENERAL ISSUES

2.1 TAKEOVERS AND THE ROLE OF EQUITY MARKETS

Efficient equity markets are fundamental to the operation and indeed existence of large companies in modern economies. The capacity for individual shareholders readily to buy and sell shares in the equity of a company increases the attractiveness of shares as financial assets, facilitates the diversification of equity portfolios and the spreading of risk and so provides the basis for companies to raise the large volumes of equity capital required for major business activities. This capacity to raise equity capital also underlies companies ability to borrow loan funds; it thus underpins the capital base for virtually the whole of the corporate sector.

The existence of efficiently functioning equity markets inherently provides opportunities for takeovers. Takeovers are just one form of market activity and action to regulate takeovers which constrains the ability of buyers and sellers to trade on terms that they regard as mutually beneficial must affect the overall attractiveness of shares as assets and thus ultimately the ability of companies to raise equity capital.

Of course, some forms of regulation can improve the efficiency of share markets and this is the basic rationale for rules such as those contained in the Takeover Code administered by the National Companies and Securities Commission (NCSC). The Code includes rules to ensure the fuller provision of information to market participants and to require offers by buyers (in specified circumstances) to be open equally to all existing shareholders. Such rules are designed to encourage 'fair play' between market participants and, in particular, to protect the interests of minority shareholders. They are not designed to assess the merits of individual takeovers, and operate irrespective of those merits. However, by providing fuller information and reducing the possibility of coercion in takeover situations, they may help market participants to make better decisions in relation to particular proposals. Such rules should be assessed in terms of their effects on the efficiency with which share markets operate rather than as a means of generally encouraging or discouraging takeover activity. Some relevant issues are considered in Section 4 of this paper.

Takeovers provide a means by which entrepreneurs who believe that they can use the assets of a company more efficiently than its existing management can bid for the company and put their beliefs to the test. The expectation that they can generate higher returns will mean that the assets will be worth more to them and they will be prepared to offer a higher price. Takeovers can thus contribute to promoting the most efficient use of existing corporate assets.

Apart from that potential of increasing the efficiency of assets in existing uses, takeovers can also assist allocative efficiency by facilitating the reallocation of capital between industries. Many firms are often reluctant to invest outside their own or closely-related industries, even though returns may be substantially higher elsewhere, as their managers' skills and experience are often highly industry-specific. Takeover specialists often have less attachment to a particular industry and are more willing to invest in alternative, potentially higher-yielding activities. In this context, firms with large cash flows operating in industries with poor to average prospects are particularly likely to be takeover targets.

The efficiency gains from takeovers do not necessarily depend on takeovers actually occurring. The mere threat of a takeover may galvanise the existing management of a company into improving its performance and raising the returns obtained on assets. It is, for example, widely believed that over the last few years the threat of takeover of certain very large companies has provided a strong spur to their managements which was not present when the companies regarded themselves as secure from takeover because of their size.

The existing managements of companies may at times devote considerable effort to resisting takeover attempts, including organising 'white knights' to purchase strategic shareholdings and arranging defensive share swaps with friendly companies. In some cases, these activities may serve no economic purpose but merely entrench existing management. If management becomes preoccupied with such defensive activities, other managerial tasks may be neglected with consequent adverse effects on the efficiency of the firm. On the other hand, the best form of defence is a high share price and much defensive activity may take forms, such as revaluing assets and rationalising and diversifying activities, which act to raise returns and increase efficiency.

The effects of takeovers on the efficiency of the corporate sector are thus considerably more pervasive than may appear from the consideration of individual instances. In that respect, they may be compared to the effects on efficiency of competition in product markets. In either case, measures taken to restrict the free operation of markets may reduce the incentives and pressures on managements to perform. Any assessment of the overall costs and benefits of takeovers needs to take this as a starting point.

2.2 THE LEVEL OF TAKEOVER ACTIVITY

There appears to be a perception in the community that takeover activity has increased in recent years. The following table provides details of takeover bids of listed companies in Australia since 1959-60 and in proportion to the total number of listed companies.

2

Year

Bids

Proportion

Year

Bids

Proportion

 
 

of Listed

 
 

of Listed

 
 

Companies(a)

 
 

Companies(a)

 
 

(per cent)

 
 

(per cent)

1959-60

60

6.3

1972-73(b)

156

10.7

1960-61

64

6.8

1973-74(b)

97

7.0

1961-62

35

3.7

1974-75

65

4.8

1962-63

42

4.5

1975-76

81

6.1

1963-64

42

4.4

1976-77

74

5.9

1964-65

44

4.7

1977-78

106

8.9

1965-66

41

4.4

1978-79

108

9.7

1966-67

44

4.8

1979-80

124

11.5

1967-68

75

8.2

1980-81

141

13.1

1968-69

83

8.7

1981-82

78

7.6

1969-70

31

3.0

1982-83

64

6.4

1970-71

100

8.8

1983-84

121

12.1

1971-72(b)

123

11.1

1984-85

121

11.4

(a) Prior to 1971-72 refers to listings on Sydney Stock Exchange only. From 1971-72 refers to total AASE listings.

(b) Excludes listed companies with debenture and loan capital only.

Source: For 1959-60 to 1971-72, Sydney Stock Exchange Annual Report 1972 and for 1972-73 onwards, Australian Graduate School of Management.

The table indicates that bids have been more numerous in the 1970s and 1980s than in the 1960s, but there is not any clear trend in bids over the last fifteen years.

The higher number of bids in the 1970s and 1980s may reflect factors such as less buoyant economic conditions in these years than in the 1960s and in particular the relative decline in the manufacturing sector in this period. These factors may have increased the attractiveness of takeovers (and improvements in the efficiency of existing assets) relative to new investment. In more recent years the increased dynamism of the financial sector may also have facilitated takeovers. It would not be surprising if the incidence of takeovers followed a cyclical pattern related to general economic conditions and there is some indication of that in the figures given above. The perception of increased activity may also reflect the scale of, and publicity surrounding, some particularly large recent takeovers. Of course, there is nothing new about large scale takeovers and care should be taken in drawing generalisations from a few cases.

3

Variations in the level of takeover activity cannot shed any light on the overall desirability or otherwise of the takeovers that have occurred. The following section reviews studies that have been undertaken of the objectives and effects of takeovers.

2.3 OBJECTIVES AND EFFECTS OF TAKEOVERS

The literature identifies a wide range of advantages which firms might seek to obtain from successful takeovers. These include:

• integration of production processes with resulting economies of scale;

• economies in administration, marketing, finance, R & D and other such activities;

• rationalisation of the use of assets and the spreading of fixed costs;

• economies through acquisition of supplying firms;

• introduction of new technology, know-how, products or designs (or application of them to other activities of the taking-over firm);

• utilisation of more sophisticated management techniques;

• replacement of inefficient managers;

• increased returns or reduction of risk through diversification;

• increased returns through reduced competition; and

• taxation advantages.

All of these offer financial benefits to the promoters and (except for the last two points) also represent gains in efficiency for the economy as a whole.

On the other hand, some writers argue that takeovers are undertaken primarily not for the financial gain of the promoters but in pursuit of non-financial objectives, such as greater prestige, power or convenience for managers in enlarged enterprises. The two sets of objectives are not necessarily mutually exclusive: in the normal course managers might be expected to seek personal advancement by successfully pursuing the financial interests of their firms. Managerial objectives are of interest only if they are pursued in conflict with the interests of the firm, or at least in ways which fail to benefit the firm.

This issue is a fundamental one, as takeovers made for objectives other than financial gain would be less likely to contribute to overall economic efficiency. Several empirical studies have been undertaken overseas and in Australia to test the motivation for takeovers and their effects.

4

A number of overseas studies, notably a study of takeovers in seven countries conducted by Mueller and others in 1980, suggest that identifiable efficiency gains from takeovers have been relatively modest overall (although they accept that substantial gains have accrued in particular cases)(1). The Mueller study concluded from this that non-financial ,managerial' objectives must predominate, although its support for this hypothesis was reached through a process of elimination of competing hypotheses and not on direct evidence that managerial objectives are significant. The recently released Australian study by McDougall and Round on the effects of takeovers on firm performance, commissioned last year by the NCSC and the Victorian Division of the Australian Institute of Management, used a similar methodology and reached very similar conclusions(2).

In a number of important respects, the methodology employed in the Mueller, and McDougall and Round studies biased the results against a finding of significant benefits from takeovers. In particular, the studies excluded from their samples firms which had undertaken more than one takeover. This removed from the samples the firms which have had most experience at takeovers and limited the studies to 'one off' cases where the acquiring firm did not seek to repeat the exercise within the given period. The studies were thus virtually confined to less successful takeovers(3).

Secondly, the studies were based on comparing the experience of the firms in the sample with a control group of matched non-merging firms. This assumes that the performance of the matched firms which were not subject to takeover can be taken as representing how the firms that were involved in takeovers would have behaved if the takeovers had not occurred. The difficulty with this approach is that the general considerations outlined above suggest that takeovers will tend to be directed to companies whose performance is open to improvement (and thus probably below average). The 'matched' firms used as comparisons were limited to firms that were not subject to takeover and thus may have been confined to better performing firms. Moreover, the management of the matched firms may have been stimulated to improve their own efficiency by seeing their competitors taken over (matched firms were chosen to be as closely as possible of similar size and industry to the firm with which they were compared). By assuming that such effects are negligible, the studies beg the questions that they are supposed to test.

---------

(1) D.C. Mueller (ed), The Determinants and Effects of Mergers: An International Comparison, (Cambridge, Mass.: Oelgeschlager, Gunn and Hain, 1980).

(2) F.M. McDougall and D. Round, The Determinants and Effects of Corporate Takeovers in Australia 1970-81, (a study commissioned by the National

Companies and Securities Commission and the Australian Institute of Management (Victorian Division) 1986).

(3) Thus the McDougall and Round study excluded such firms as IEL, Bell, Adsteam and FAI which have a record of successful takeovers and strong growth in share values. According to a recent study by the Sydney Stock Exchange these companies were among the best performing stocks of the past decade. An initial investment of $1,000 in any one of these companies on June 30 1975 would have yielded over the following ten years between $70,000 and $178,000 (Personal Investment, February 1986,

p. 26).

5

Thirdly, extraordinary gains and losses have been excluded from the profit measures used in the studies. This is likely to result in significant underestimation of the gains achieved in many mergers, as rationalisation of operations and disposal of surplus assets are likely to be important motives in a significant number of acquisitions.

In view of these methodological problems, which have also attracted comment from academic sources, it is not surprising that these studies could find little benefits from takeovers. It is however notable that, even so, the Australian (McDougall and Round) study found that:

• the growth rate of the merged firms was significantly greater after takeovers than their components were before;

• the growth rate of the matched (comparison) firms also increased after the takeovers (which is consistent with the matched firms receiving some stimulus from the threat of takeover); and

• shareholders in target firms gained significantly from the takeovers, with much of the gain occurring prior to takeovers (which the study attributed to market anticipation of takeovers and premiums paid in the prices in takeover offers).

The McDougall and Round study also indicated that post-takeover returns to shareholders in acquiring firms were less than those in the matched firms, from which the study concluded that shareholders in acquiring firms did not gain from takeovers. This contrasts markedly with the share market experience of specialist takeover firms (see footnote 3), which were of course excluded from the sample. No attempt was made in this study to measure the net gains or losses of all shareholders (both in target and acquiring firms) involved in takeovers over the period of analysis, although the study did note the results of other studies (both in Australia and overseas) that takeovers appear to create excess returns to the shareholders of participating companies and that there is a remarkable consistency in these results.

As regards the motives underlying takeovers, the McDougall and Round study tested only two possible motives (economies of scale and risk reduction) and its findings in these areas were inconclusive. It argued that economies of scale were not significant on the grounds that the results showed that acquiring firms were larger than their targets. However this argument is illogical since the gains from economies of scale may well be most marked when a small firm combines with a larger one.

The literature as it stands at present is therefore somewhat inconclusive - though, perhaps, unjustifiably so. It may nevertheless be observed that, whatever the immediate motivation that managers may have in individual takeovers, in a competitive market place they will be under pressure to obtain returns that justify their acquisitions. Particularly in the case of major takeovers where very large sums are involved, it seems unlikely that the owners would accept the very considerable risks involved without some prospect of financial gain merely to support the comfort or prestige of their managers. Any conflict between managerial and financial objectives may therefore be much less significant in firms undertaking major takeovers than in less dynamic firms where there may be less cause and opportunity for the activities of managers to be subject to close scrutiny by the firm's owners. Certainly it has generally been assumed in public debate in Australia that recent major takeovers have been undertaken with financial motives.

6

While no systematic study has been made of the outcomes of foreign takeovers examined under foreign investment policies, the experience of the FIRB suggests that actual results rarely match the claimed benefits and expectations of the parties, as expressed at the time of seeking approval. It is of course not at all surprising that applicants arguing a case in attempting to obtain approvals should portray their proposals in an optimistic light. Similar difficulties would confront any wider procedures for vetting, the merits of takeover proposals. Less than expected outcomes may also have been influenced by depressed economic conditions.

Even if it were established that many takeovers have not produced large benefits, those takeovers could still have been economically worthwhile, so long as their benefits exceeded their costs. Thus across-the-board measures that discourage takeovers generally, or make them more difficult, may have a net adverse effect if they prevent those takeovers that would have had net benefits. Moreover, the available studies do not attempt to measure the efficiency effects induced by the threat of takeovers on companies that have not been taken over.

On the other hand measures to screen proposed takeovers selectively in terms of their expected benefits are not easy to operate. The benefits of individual takeovers may not always be readily identified in advance and outcomes are not necessarily in line with expectations. That is hardly surprising: circumstances change, the capacity of new managers may not be as good as their promises and the information originally available may have been defective.

The fact that some poor decisions are made is not a sufficient reason for regulating takeovers. Any investment decision may turn out to be less than first hoped. Markets tend to exercise their own disciplines in such cases. Thus, in many takeovers, the past track record of a potential bidder in undertaking previous successful takeovers is likely to be an important consideration in the bidder's ability to obtain backing for bids. Those who do not achieve rates of return sufficient to cover the prices they have paid are unlikely to prosper.

3. SOME IMPLICATIONS

3.1 EFFECTS ON COMPETITION

In some cases, takeovers may result in reductions in competition, thereby allowing prices to be raised at the expense of the consumer. In these cases, the resulting private gains to the companies would not represent genuine improvements in efficiency for the economy as a whole.

7

The Trade Practices Act already provides for the scrutiny of mergers and takeovers where they may result in a reduction in competition. It also includes provisions to limit abuses of market power. The Trade Practices Commission (TPC) has scrutinised a number of recent takeover bids involving firms in similar fields. In some cases, it has sought and obtained undertakings from bidders as to future conduct should their bids succeed (eg Bond/Castlemaine Tooheys and Amatil/Fielder Gillespie). In other cases, the TPC has expressed concern at the possible anti-competitive effects of the proposed merger and indicated that it will monitor post-merger developments closely (eg Coles/Myer merger). According to a recent press statement by the TPC Chairman, the TPC only institutes court proceedings if there is no feasible alternative available (eg Fielder Gillespie Davis takeover proposals for Allied Mills and Goodman Group of New Zealand). Such cases can be protracted and the outcome uncertain because of the fine judgments required (and the TPC carries the onus of proof).

Whether the TPC's powers to intervene in relation to takeovers which it regards as anti-competitive are too limited is a matter which requires careful judgment. The cases mentioned above certainly have elements which justify concern, but it is not clear that the approach adopted by the TPC, of monitoring developments, will prove inadequate. Assessments of anti-competitive actions are complicated because, while a takeover may result in reduced competition within an industry, there may be offsetting benefits in the form of scale economies and avoidance of fragmentation. Also, so long as there are no artificial barriers to entry to the industry, market dominant firms should be constrained from excessively anti-competitive practices. Increased regulatory action, particularly where this is pursued through extended legal processes of litigation, may be more disruptive than the problems it is intended to remedy. In any case, to the extent that further action is warranted, the appropriate course would be to strengthen the relevant specific provisions of the Trade Practices Act rather than to take action to limit takeovers generally.

Although there has been a long history of takeover activity in major countries over the last century or so, there seems to be little indication that this has led to overall decreases in the intensity of competition over the long term. In part this may reflect the effects of anti-trust or trade practices legislation. In many areas it reflects increased international competition where barriers to trade have declined. But it probably also reflects some underlying limitations on corporate growth: large organisations tend to develop their own internal problems of maintaining managerial control and efficiency. The entry of new firms may then bring new competition, and technical innovations can also help to break down existing concentrations of market power.

3.2 TAXATION ASPECTS

Concern has been expressed in some quarters regarding the possible cost to revenue of takeover activity - that cost deriving from tax deductions allowed to the predator company.

For the most part, that concern has related to takeovers where the predator uses a high level of borrowings. Some however, have also expressed concern with the opposite situation; ie predator companies which resort to equity - through the issue of redeemable preference shares - rather than debt to gain tax advantage.

8

This dual concern with the loss of tax revenue from two opposing situations - 'excessive' resort to either debt or equity - indicates a need to look to the principles involved and cautions against hasty adoption of 'solutions'.

Similar caution is required in responding to the claim that takeover activities involve the exploitation of tax loopholes which must be closed. Where loopholes exist they should, of course, be closed - but it is always essential to ensure that a loophole does, in fact, exist. There will be occasions, for example, where the tax advantage derives from a justifiable reaction to unintended or undesirable features of the tax law or its interpretation. In other cases, competing principles may be involved or the tax system may be being expected to serve competing objectives. In some cases, 'solutions' which yield the highest revenue may be quite contrary to basic economic or social objectives.

The apparent paradox referred to earlier - that there may be tax advantages (and revenue losses) associated with both high gearing and low gearing - is no more than a reflection of a fundamental design feature of the classical income tax system; namely, that debt and equity payments by a company are treated differently:

• debt payments (interest) are a deductible expense and hence will reduce tax liability; and

• equity payments (dividends) are not a deductible expense, being paid out of after-tax income, and will have no direct effect upon tax liability.

It follows that:

• companies with high levels of taxable income may see tax advantage in debt rather than equity finance - ie in a shift to higher gearing; and

• while companies with low levels of taxable income (or in tax loss) will see no such tax advantage, but may see cash flow advantages in a shift to equity financing or lower gearing.

In both situations, the tax system is operating as intended. Moreover, in both situations, commercial judgments on factors other than tax are involved and will frequently dominate.

That is not, of course, to say that there are no cases where tax considerations are paramount. Nor is it to say that the tax system is neutral with respect to the debt/equity choice. The present tax system is not neutral in that regard, and that is an important focus of some of the main elements of the 19 September 1985 tax reform package.

What it does indicate is that strong arguments would be needed to justify action to restrict high gearing - by, for example, restricting the deductibility of interest payments. That would be particularly the case in respect of takeovers as it is a fact that certain takeovers - such as that of a large company by a smaller one - cannot occur without increased gearing by the predator, at least for a time. Measures which restricted such takeovers could make some companies immune from takeover - other than by other large companies or, particularly, overseas interests.

9

Moreover, if it were decided to change taxation arrangements which figure prominently in the case of takeovers, or highly geared transactions more generally, it would be equally important to be sure that such arrangements could be made to work effectively. Some inescapable facts regarding the operation of capital markets suggest that this is unlikely to be the case. Experience both here and overseas indicates that legislation which attempts the impossible has the capacity to do more harm than good.

The following sections examine these issues more fully, examining the sources of the revenue loss associated with highly-geared transactions, identifying some non-neutralities in the present tax system and indicating measures proposed to ameliorate them, and canvassing other measures that have been suggested to limit revenue costs.

Gearing and tax revenue

Any increase in borrowing by a tax-paying company, for whatever purpose, is likely to involve a potential cost to revenue. That is because the interest payment by the company will involve a cost to revenue at the rate of 46 per cent (the current company tax rate) while the gain to revenue from taxation of the corresponding interest receipt may well be at a lower tax rate. The revenue gain, for example, could be at any of the following rates:

• 60 per cent if the borrowing is from an individual paying the top marginal rate of personal tax;

• between 48 per cent and zero if the borrowing is from an individual paying marginal tax rates below the top rate;

• 46 per cent if the borrowing is from another taxpaying company;

• 10 per cent if the borrowing is from an overseas source subject to interest withholding tax (IWT); and

• zero if the borrowing is from an overseas source not subject to IWT, or from a tax-exempt domestic source (such as superannuation funds or State Banks).

Although there is the possibility of a gain to revenue from some of these routes, in the generality of cases an initial revenue cost will be involved. Moreover what matters is the ultimate lending source so even if the borrowing is from a resident company taxed at 46 per cent (such as a taxable financial institution), if, for example, that company borrows in turn from overseas then a revenue cost will still ensue.

The following points should be made regarding this potential cost:

• first, the cost arises in any borrowing for income-producing purposes; it will arise in the case of borrowings for takeover purposes but obviously is not restricted to such borrowings;

10

• secondly, the revenue cost is only a potential cost; a direct revenue cost is involved in the borrowing transaction, per se, but whether there is an ultimate cost to revenue depends on what other transactions are associated with the borrowing and what transactions the borrowing might displace; this aspect is discussed further in Attachment A;

• thirdly, the cost arises from different rates of taxation of interest receipts; the reasons for those differences are discussed further below; and

• finally, the revenue cost is exacerbated because the income tax applies to nominal, rather than inflation-adjusted, interest payments and receipts.

Notwithstanding these points, there is a question as to whether the tax system encourages borrowing by corporations. This is certainly the case with the present classical system of company taxation which, so far as taxable shareholders are concerned (though not exempt institutional shareholders), involves the double taxation of dividends and constitutes a marked bias towards debt rather than equity financing.

This bias between debt and equity financing arises at the company level:

• the individual investor is taxed equally on receipts of dividends or interest - there is no bias there; and

• at the company level, however, interest payments are deductible in the calculation of taxable income, thus reducing company tax; dividend payments, however, being paid out of after-tax income, are not deductible and corporate income financed by equity is thus subject to taxation at both the company and individual levels (double taxation).

Imputation

The imputation system of company taxation, to operate from 1 July 1987, will end the double taxation of dividends and hence ameliorate substantially the bias towards debt financing. This will be achieved by providing resident individual shareholders with a credit, which may be offset against their personal tax, for tax already paid at the company level on income giving rise to dividends. All other things equal, debt financing, including in respect of takeovers, will be reduced under the imputation system. In the case of tax-exempt institutional shareholders, who are not now subject to double taxation, there will, of course, be no change.

Because the classical system involves the double taxation of dividends, it also tends to encourage the retention rather than distribution of company income. Large accumulations of undistributed income are not always reflected in share prices and can tend to attract takeover interest. Moreover, companies with high retentions, and limited investment opportunities in their own field, may seek takeover targets lest their own company be subject to a similar fate. By removing this bias towards retentions, the imputation system can be expected to also reduce this attraction to takeover activity generated by the classical system.

11

Under the imputation system, dividends paid after I July 1987 will attract imputation credits for dividend receipts and will also be subject to payment of compensatory tax by the company. That compensatory tax can be offset against company tax paid from that date which, in the first instance, will be tax in respect of income earned in 1986-87. For that reason, some of the desirable effects of imputation will be apparent in 1986-87 or, in the cases of some companies with substituted accounting periods, somewhat earlier.

The encouragement provided under our pre-19 September company tax system to realise capital gains from the sale of target companies' assets (in the absence of a capital gains tax, the gains, on distribution, would generally be taxed only once as opposed to twice with distributions of trading income) will also be addressed directly by the imputation system. Under imputation, and through the compensatory tax effect, distributed capital gains will be taxed once, equivalently to distributed trading profits.

Capital gains tax

The absence of a comprehensive capital gains tax (CGT) prior to 19 September 1985 created an incentive for corporate transactions and financial policies that produced capital gains rather than dividends for shareholders. An asset-rich target had the potential of yielding tax-free gains from the sale of assets after the takeover or on any gains from the subsequent disposal of the target company's shares. The recent introduction of the CGT will ameliorate this tendency.

Regarding the more prominent corporate raiders, it may be thought that the new CGT is of little relevance, on the grounds that they would have been classified as share traders and already subject to tax on their capital gains under section 25 of the Income Tax Assessment Act. However, that is far from certain. In individual cases, the facts were often presented so as to give a picture of a transaction that was on capital account and the outcome frequently required litigation.

A particular feature of the US and Canadian tax regimes that has been criticized as representing a 'tax subsidy' for takeovers is the rollover provision in the capital gains tax regime of both countries for share-for-share exchanges in takeovers. Our CGT regime does not provide a rollover in such circumstances and hence such a tax subsidy for takeovers will not be present. The significance of this will increase over time as an increasing proportion of the total stock of shares becomes 'ungrandfathered' and subject to CGT.

Group loss provisions

The group loss provisions are generally available provisions introduced to reduce constraints in the law on obtaining the benefit of current tax losses. However, they only apply to 100 per cent common ownership companies. They are highly desirable on tax principle grounds (organizational form should not affect a corporation's aggregate tax liability) and do not appear to be at the heart of the current concerns with takeovers, most of which, being partial takeovers, would not bring the group loss provisions into play. Where 100 per cent takeovers are completed, tax benefits can be taken through the resultant ability to offset tax losses within a group; however a similar result may be achieved, in any case, by restructuring the group.

12

Redeemable preference shares

The Treasurer announced on 7 April 1986 that redeemable preference shares (RPS) with an effective term to maturity of two years or less, and which are issued after that date, are to be generally treated as debt for income tax purposes, ie 'dividends' on such shares will not be eligible for the section 46 rebate but will be taxable income of the lending company and deductible to the borrowing company. The imputation system is likely to largely eliminate the tax advantages attaching to longer-term issues of RPS because compensatory tax would be payable when preference share dividends were paid to shareholders and, in the case of many current issuers of such shares, there would be insufficient company tax liability against which it could be offset. These factors are relevant to the use by companies of RPS as a financing mechanism including in a takeover context. RPS had tended to be utilized fairly heavily by corporate raiders, who frequently had substantial rebatable dividend income and relatively little taxable income against which interest outlays could be written off.

In brief, the post-reform tax system would appear to address the major flaws in the old system which might be regarded as being associated with takeover activity.

Policy options

Notwithstanding that conclusion, tax changes have sometimes been suggested to raise the cost of highly geared takeovers. Although such changes are not considered to be warranted, or desirable, on account of takeovers, the following discussion of the main options is provided to highlight the further issues and problems they involve. The main options are:

• tax indexation;

• changes in interest receipts taxation;

• thin capitalization rules; and

• quarantining of interest payments.

Indexation

As indicated in the draft White Paper on Reform of the Australian Tax System, comprehensive indexation of the income tax system (including deductibility of real, not nominal, interest) would be a desirable measure in its own right. However, it represents a very major change which would require lengthy developmental work and raises issues well beyond the takeover area. Net revenue losses of the order of $1-2 billion per annum could be involved, depending on the rate of inflation.

13

Partial indexation confined to taxation of real, rather than nominal, interest is the most difficult area in development of comprehensive indexation measures, and would still be a massive exercise. Partial indexation would introduce some distortions and there could be a large ($400-600 million per annum) cost to revenue. That cost would arise because interest on some large slabs of debt (eg borrowing by governments and for home-ownership) does not attract tax deductibility; therefore, in aggregate, indexation would affect the taxation of more interest receipts than interest payments. It is not, therefore, considered a viable option.

Changes in interest receipts taxation

There are major constraints, including those relating to double tax treaties, on increasing the IWT rate. The purpose of such treaties is to agree on taxation arrangements between the two countries involved and so avoid over-taxing (so-called double taxation) of income flows. In the case of interest payments, taxation by the country of source is limited (to 10 per cent for most treaties) while taxation in the country of residence of the lender depends upon that country's tax system and rates. The main area for manoeuvre would be to cut back IWT exemptions, especially those relating to public or widely-spread issues of securities. The basic exemption was introduced in 1971 (and then related only to 'bearer' securities) in conjunction with the establishment of the AIDC, as a means of dispensing with the need for companies, for technical reasons, to channel borrowing on the European market (where 'bearer' securities are the rule) through a subsidiary in a country which did not levy IWT. In 1983, the exemption was widened beyond bearer securities because of difficulties with registered securities being subject to IWT, which had the effect of discriminating in favour of European rather than US financial markets. Cutting back on IWT exemptions would be against the trend in most countries.

Taxing domestic exempt institutions (eg superannuation funds) on their interest receipts would represent a fundamental change in an area where the Government has already moved decisively in the opposite direction (ie to tax lump sum superannuation proceeds more adequately in the hands of individuals). Moreover, the development of occupational superannuation now represents a cornerstone of the Government's retirement and wages policy, having been negotiated in the context of the Accord Mark II.

There could be constitutional problems in seeking to tax the exempt State banks.

Thin capitalization rules

At present the use of 'thin capitalization' - high debt and low equity - by foreign investors in Australia as a means of tax avoidance is controlled, in certain circumstances, by FIRB approval processes in cases which are examinable. The circumstances are those where funds are borrowed from foreign non-arm's length sources (typically, the foreign parent company). In such circumstances, the Government seeks to ensure that the capital structure employed, and especially the proportion of debt to equity funds employed, reflects commercial practices. If excessive amounts of foreign investment came in as debt capital, the income of the Australian company could be largely freed from company tax

14

because of the large amount of deductible interest, and the only Australian tax on it would be IWT at a rate of 10 per cent. Domestic corporate tax would be escaped by, in effect, paying profits out as interest payments. Therefore, where funds are borrowed from non-arm's length sources, the Government generally requires that the proportion of foreign debt funds to foreign equity funds not exceed a ratio of 3:1 (6:1 for financial institutions). Many countries have such rules for foreign investors as part of their income tax law.

The requirement does not apply where the borrowing is from genuine arm's length sources, for two reasons. First, there is not the same incentive in such cases to convert profits into interest payments. Secondly, any reputable lender would usually require that there be a reasonable level of equity in the domestic subsidiary, and that a commercial rate attach to the lending. The requirement is also not necessary where the funds are provided by another Australian company; in this case, the interest deductible by one company is taxable income of the other.

Hence, the existing thin capitalization rules represent a common anti-avoidance provision relevant to non-arm's length funding from foreign associates. They therefore have very limited relevance to a measure which would apply across-the-board to all domestic companies, covering arm's length and other funding.

An across-the-board measure would presumably specify either a single debt/equity ratio, or several ratios depending on the industry. The measure could specify that any interest payments attributable to debt in excess of the relevant ratio would be disregarded for tax purposes, so the interest payment would not be deductible.

No country employs such an across-the-board rule. Rules to impose a gearing ratio or ratios on companies would have the following undesirable effects:

• any single ratio would be too high for acceptable standards in some industries and too low for others;

• attempts to set multiple ratios for the full diversity of industries would involve government agencies making judgments of essentially a commercial character without the necessary experience of the sectors or industries involved; and

• companies which might be geared below a permissible ratio for commercial reasons would be encouraged to gear up to gain the tax advantages of doing so and to guard against takeover. The measure, therefore, could be self-defeating.

It should also be noted that limits on the gearing of companies undertaking takeovers:

• would protect large companies against takeovers by smaller firms, since such takeovers necessarily depend heavily on borrowed funds (at least for a time), but not small companies against takeovers by larger ones; and

• depending on how the limits were applied, could disadvantage Australian companies relative to foreign companies in undertaking takeovers. The foreign company could achieve higher overall gearing by increasing its borrowing in its home country and claiming deductions there. It would make takeovers easier for foreign companies (and possibly foreign subsidiaries of Australian companies) and would be open to criticism on that score.

15

In order to avoid undue disruption, it would be necessary to include some transitional provisions to enable highly geared companies to adjust. Any such transitional provisions would discriminate against new firms; that seems unavoidable. There would also be hard definitional questions as to the coverage of both debt and equity.

The measure is considered to be highly non-neutral - it would hit hard at companies forced to borrow heavily during hard times - and is not recommended.

Quarantining of Interest Payments

There are various possible quarantining propositions involving restrictions on the deductibility of interest payments.

One such proposition that has attracted some attention arises from an alleged 'loophole' in the operation of sections 46 and 50 of the Income Tax Assessment Act. The basic purpose of section 46 is to ensure that dividends, on which company tax has already been paid, are not subject to further taxation as they pass through the corporate chain before being taxed again (under the classical system) in the hands of the individual shareholder. The alleged loophole is that interest expenses incurred by one interposed company in purchasing the shares and earning the dividend income is not offset against the dividends in calculating the section 46 rebate.

The issues involved are rather complex and no doubt complicated by the fact that the wording of section 46 derives from the period before Australia adopted the classical system of company taxation; and the wording of section 50 derives from a period when Australia taxed different forms of income at different rates. Much of the concern on the general issue may arise from a misinterpretation of the present intent of those sections.

The approach suggested to deal with the alleged loophole would involve action aimed at seeking to ensure that interest on finance borrowed to acquire shares be offset against the rebatable dividend income resulting from it. A somewhat similar idea was considered in 1971, when the then Treasurer announced that the Government would amend the law so as to ensure that a share of all allowable deductions (including interest) was offset against dividend income of companies before calculating the section 46 rebate on it. After examining the matter closely, the then Government dropped the proposal, in light of the fact that it had 'become quite clear that there is no simple way of altering the basis of the section 46 rebate without giving rise to serious anomalies between different company structures'. The general issue, including the 1971 episode, is discussed at Attachment C. That discussion, which covers issues of principle and practicality, is summarized briefly below.

On conceptual grounds, expenses incurred in earning assessable income should be deducted from that income in determining taxable income. Income, and now capital gains, are taxed when realised by companies. That after-tax income may be passed through a chain of companies where section 46 - which provides resident companies with a tax rebate on dividends received - operates to ensure that no further tax effect applies until the income is received by individual shareholders. The approach referred to in the preceding paragraph would seek to reduce or eliminate that rebate where interest expenses had apparently been incurred in earning the dividend income.

16

The crux of the problem here derives from the fact that money and related debt or equity instruments are fungible. That fact cannot be ignored at the practical level. Legislation that 'attached' certain debt to certain income would require tracing rules that could be avoided by all but the uninformed or imprudent. It would be a straightforward matter to structure arrangements such that all borrowings would be attributable to taxable income and all equity ascribed to non-taxable (rebatable or exempt) income.

The practical problems caused by the fungibility of money help throw some light on the issues of principle; ie the practical difficulties arise because, in principle, attributing interest expenses to particular revenue items has no economic rationale. It is simply not possible, on economic grounds, to say that a particular interest expense relates to a particular item of income. It matters not that, as a matter of historical fact, certain debt was incurred at the same time as certain revenue - or the future right to that revenue - was acquired. Legislation which attempted to ignore this principle would quickly run foul of the practicalities, referred to in the preceding paragraph, which merely flow from the basic economic facts underlying the principle.

Apart from questions of principle and practicality, there are also those relating to economic effects. Some evidence on this last aspect is provided by a similar quarantining measure which operated in Canada before it was abandoned in 1972 (see Attachment D, which also contains commentary on the situation in the US and UK). The provision was repealed for three main reasons:

• it put Canadian corporations at a gross disadvantage compared with foreign corporations in bidding for Canadian corporations, as the foreign corporation could claim deductions for the interest in its home country;

• it was only effective against hostile takeovers, as friendly ones could be arranged by purchase of assets rather than shares; and

• determining which funds financed the takeover and which financed some other activity was a difficult matter, particularly where the acquiring company was large. The effect of the tracing rule used was to give an advantage to large over small corporations in takeover bids.

As suggested by the discussion above, the same arguments would apply with equal force in Australia today.

A more general quarantining measure that has been proposed would involve restricting the deductibility of interest on funds used to buy shares up to the amount of the dividends and other taxable gains derived from them.

It has been suggested that such a measure could be justified on the same grounds as the quarantining proposal for rental property (which, incidentally, applies consistently to, rental property' companies, trusts and partnerships as well as individual taxpayers). However, in the case of negative gearing of rental properties, there was a serious mismatch between the time at which deductions, including nominal interest deductions, for expenditures producing income were allowed and the time at which income was brought to account, substantial income from rental property traditionally being in the form of capital gains either tax free or, as proposed, taxed on an indexed, deferred basis. In general, this is not the case with normal corporate activity; the interest expense is written off against current income which is taxed contemporaneously.

17

Further, the effects discussed earlier of the imputation system with its compensatory tax on distributions from capital gains realised by companies - say through takeover activity - set companies somewhat aside from the case of other taxpayers realising preferably taxed capital gains (although companies need to be included in the scope of the quarantining measure for rental property for obvious anti-avoidance reasons). In contrast to the treatment of capital gains realised generally, the capital gains underlying company dividends will be taxed, through the effect of compensatory tax, equivalently to distributed trading profits.

In assessing such a proposal, a lot would depend on how it was determined whether borrowed funds had been 'used to buy shares'. If a parent company injected funds which it had borrowed into an existing subsidiary through an equity issue by the latter, it would seem to come within such a test. Yet it would still be the case that the interest expense was being matched by current taxable income to the same extent as if the subsidiary had borrowed the money directly. With or without attempting to allow for such arrangements, the attempt to ascribe a purpose to a particular borrowing would pose the same massive conceptual difficulties mentioned above.

Apart from lacking a firm conceptual basis, a quarantining measure in respect of share purchases generally would also suffer from most of the practical drawbacks identified with the previous option, as well as some others. It would be readily avoided by all but the most poorly advised investors.

There is also the question of the scope of any such measure. The rationale for any action in this area derives from revenue considerations, which in turn derive from high gearing (and the associated interest receipts tax arrangements), not from takeover activity. Viewed in this manner, there is no case for confining any quarantining measure to takeovers. A measure confined to takeovers would discriminate against businesses acquired in this manner; highly geared investment through takeover would pay higher company tax than highly geared companies in general. Also, as noted above, any measure confined to takeovers would be easily circumvented in many situations, by the takeover company transferring profitable assets to itself. If the proposal were meant to be directed at the more prominent corporate raiders, it is essentially absurd.

A measure which embraced all share purchases could also be fairly readily circumvented, by substituting arrangements to negatively gear the acquisition of the target company's income producing assets instead of its shares. This option would not be available to all, of course; 'unfriendly' takeover activity would be made more difficult, and to that extent the measure would be highly discriminatory.

If one wished to quarantine interest arising from financing of 'friendly' takeovers of income-producing assets of target companies, that would pose a question as to how such transactions are to be distinguished from completely arm's-length investments. The only tax effective stopping point would appear to involve applying the measure to all income producing assets; ie seeking to identify what was done with each and every borrowing, and limiting interest deductions to income from that deployment of the funds. Given the very wide variety of transactions to which it would apply, it would inevitably introduce its own distortions, including in situations where the owners of income producing assets were in financial difficulties and forced to borrow heavily. Any such approach would be massively damaging to the Australian economy.

18

Conclusion

It would appear that a lot of the heat in this issue has been generated by concern about takeovers, per se, and, for the rest, by misinterpretation of the intent of some tax legislation. The latter may owe something to the wording of legislation designed for another purpose and now outdated in relevant respects; it may also owe something to a disregard of economic principles and a lack of attention to the history of the matter.

The discussion above - and in the attachments - indicates that the issues are more complex than some of the public debate has acknowledged. As for the tax aspects that have gained prominence in the debate, they all relate to general features of the law; in most cases, the tax advantages to the acquiring company are also available to the target company, and usually have nothing to do with takeovers. Some of the perceived problems will be largely, if not wholly, corrected by the reform measures now in train. Tax changes proposed to be confined to takeover situations would be easily circumvented. If unnecessary disruption to equity markets is to be avoided, it would seem sensible to wait for the reform measures to run their course.

3.3 GEARING

Beyond the concerns relating to the perceived tax advantages of high gearing, discussed above, there are other concerns about the effects of takeovers on gearing, per se.

While the adoption of more highly geared financing arrangements increases the risks attached to the equity component of that financing, the greater the risk, the more difficult it will tend to be for the firm to raise additional equity capital. Firms need to set a balance in their reliance on equity and debt in the light of their assessment of the various other risks affecting the operation of their business and its future prospects. The appropriate degree of gearing in any particular situation is thus essentially a matter for commercial judgment. Gearing will normally vary from firm to firm and between different industries. There is no single guideline which would be appropriate in all circumstances - indeed any single guideline would inevitably be inappropriate to the circumstances of many firms.

Increasing sophistication in financial management and changes in commercial practice may have affected attitudes to gearing in many areas of business in recent years, but there is no indication that this has been a source of instability to business overall or that there is reason for general concern in this regard. Increased gearing, provided it is based on sound commercial judgment, may be seen as allowing more effective use to be made of available equity funds.

In the case of a highly geared takeover, the financial institutions that underwrite the bid (and, indeed, the bidding firm itself) presumably consider the resultant gearing acceptable. In some cases, where takeovers have involved particularly high levels of borrowing, this has been seen as a short term arrangement and the takeovers have been followed by asset divestments or share issues to lower the debt/equity ratio. There seems little indication that high gearing promoted by takeovers has involved unsound financial practices or led to financial difficulties.

19

A related concern is the possibility that a firm with a high level of gearing as a result of a takeover will be less able to finance new investment in view of its debt servicing obligations. This concern may derive, in some cases, from an undue preoccupation with internal sources of funds as a basis for investment. Alternatively it may reflect a view that the investment concerned may not generate rates of return sufficient to cover the cost of raising the necessary funds externally, whether through additional equity finance or borrowing. In either case, the concern may be seen as reflecting inefficiency in the investment practices of the target firm prior to takeover, which the takeover in effect would be remedying. The difficulties involved in attempting to impose a constant gearing ratio (or ratios) on companies were noted earlier.

3.4 DIVERSION OF FUNDS FROM PRODUCTIVE INVESTMENT

A further criticism made of takeovers is that they may divert financial resources away from new productive investment into speculation on existing assets. There is clearly an element of speculation in many takeover situations. That reflects the risks and uncertainties involved. That speculation, however, is based on expected increases in the earnings of firms when they have been taken over. Speculation is thus a counterpart to the productive or other financial gains expected to flow from takeovers; indeed it may provide a stimulus for seeking those gains. Speculation and increased production are not necessarily at odds in this context; they may go hand in hand.

Nevertheless concerns have been expressed that the diversion of funds into financing takeover transactions may reduce the funds available for new investment. In assessing this criticism, account needs to be taken of the funds released by a takeover as well as those put into it. Some of the funds released by the sale of shares in a takeover may result in increased consumption but, as no obvious change in overall saving propensity is involved, it is likely that the bulk will be used for purchases of other existing assets, the repayment of debt or new investment. Purchases of other existing assets will in turn release further funds for other uses. It will take some time for such series of transactions to work themselves out, but in the longer term the main effect is likely to be a reallocation of the ownership of debt and equity between firms rather than changes in the total volume of funds available for new investment.

There appears to be little direct evidence on the impact of takeovers on new investment. However, as already noted in the section on objectives and effects of takeovers, the available data suggest that merged firms grow faster after takeover than their components did before (even when the sample excludes the more successful takeovers). This suggests that takeovers, by promoting more efficient and aggressive management, have positive effects on the dynamism of firms. In this context it seems likely that the overall volume of new investment is stimulated by takeovers rather than depressed.

3.5 EXTERNAL ACCOUNT ISSUES

Much of the financing of some recent takeovers (or takeover bids) has been sourced from overseas borrowings. This is, of course, not unusual given the high degree of

20

integration of Australia into world financial markets. Since the gross flows involved in such transactions may be substantial, this has led to concern about possible impacts on the balance of payments, external debt and the exchange rate. Regardless of their magnitude, individual transactions or groups of such transactions do not necessarily have any implications for aggregate net national borrowing - outflows are also involved and consideration needs to be given to the consequentials of specific flows.

The overall external account implications turn not on individual gross flows but on whether any specific incentive or disincentive to takeovers in general affects the overall relative attractiveness of lending to Australia or borrowing from abroad. If there are forces at work which make takeovers of Australian companies particularly attractive, that is likely to have implications for the shape of Australia's external accounts. That, in itself however, does not rouse any new concerns beyond the 'domestic' ones.

The immediate effect of inflows of funds to finance takeovers is to add to external debt. However, the longer term effects depend on how those funds are used and what consequential transactions occur. Funds released by takeovers may be used for the repayment of existing overseas debt, for purchases of overseas assets by Australians, or for lending to domestic borrowers in a manner which displaces borrowings that would otherwise have been made from overseas. Such transactions operate to offset the initial inflow of funds. Any impact on the balance of payments, external debt and the exchange rate of an initial borrowing transaction to finance a takeover will thus tend to be muted by subsequent consequential flows in the other direction.

The overall impact which gross overseas borrowings for takeovers may have on the balance of payments and overseas debt has to be assessed not in terms of individual financial transactions, but of possible influences on total borrowings in the economy and, related to that, total investment. Takeovers may have contributed to a shift to higher gearing (though there is no concrete evidence to that effect) and thus to greater total borrowings in the economy, and to a larger component of borrowings in capital inflows.

In this respect, it is not apparent that borrowings undertaken overseas to finance takeovers are different in their external account consequences to overseas borrowings for other purposes. The servicing of any increase in net debt resulting from takeovers, or other borrowing, will be forthcoming if the borrowing has been well judged. The fact that a particular borrowing is applied to a takeover does not bear in any specific way on that question.

That is, the significance for the balance of payments of an overseas borrowing for a takeover is no different from that for overseas borrowings for any purpose. Initially overseas borrowings may put upward pressure on the exchange rate. Such upward pressure will be countered by downward pressure resulting from outflows to service the overseas borrowing. (The timing of these influences would be affected by possible hedging arrangements.) If the returns on the use of the capital are greater than the costs of servicing the borrowing, as would be the longer run presumption, then this will contribute as it has in the past to higher national income supported by the pattern of the external accounts whereby Australia experiences a net capital inflow matched by, a current account deficit.

Future servicing commitments resulting from takeovers are in principle provided by the future returns on those investments. Such rates of return may of course be influenced by

21

government interventions, especially the taxation system. If takeover activity is non-neutrally affected by the set of government interventions bearing on after-tax rates of return, then the activity will be potentially encouraged. If this encouragement produces a relatively more attractive investment climate in Australia than abroad (other things equal) this will produce consequences for the external accounts. This does not mean, of course, that from the particular perspective of the balance of payments, exchange rate or total overseas debt and equity obligations, overseas financing of takeovers raises additional specific concerns beyond the 'domestic' ones.

Special considerations arise in connection with takeovers by foreigners, where there may be political and other objections to substantial increases in foreign ownership and control. However, appropriate legislation and procedures are already in place for assessing such cases and we see no reason to suggest that they are not working adequately.

4. SOME REGULATORY ISSUES

4.1 THE ROLE OF CORPORATE RAIDERS AND INVESTMENT FUNDS

Resales of shares have been an important element in a number of recent Australian takeover bids. Corporate raiders have been able to make substantial profits at the expense of a target company or its defenders by assembling a strategic shareholding, threatening to take control and then selling their holdings at a higher price. In many cases it appears that the takeover attempt has been genuine and not primarily motivated by the prospect of such profits.

Whether such actions have represented genuine takeover attempts, or have been undertaken with a deliberate view to resale, they may still serve a useful economic function. They provide a signalling function in identifying potential takeover situations where assets are underutilised. Failed takeovers are very frequently followed by a successful takeover of the target company within a relatively short period and the management of a target company is given a very clear stimulus to improve its performance.

It has also been argued that the changed role of institutional funds managers has contributed to the relative ease of mounting a takeover in recent years. Because of greater competition between fund managers, as reflected in the regular publication of comparative performance reports, institutions are now more prepared to sell to takeover bidders (thus realising, short-term profits) and even to build up strategic holdings of shares in anticipation of bids. Similarly, the substantial deregulation of the stockbroking industry has put considerable pressure on brokers' margins prompting some to take an active role in assembling parcels of shares for potential bidders, at times by buying shares on their own account. While this activity by institutions and brokers is often interpreted as precipitating bids which might not otherwise be made, it may be more accurately viewed as representing intermediation in the takeover process, and thus increasing the efficiency with which it occurs.

22

4.2 DEFENSIVE MEASURES BY TARGET FIRMS

As noted earlier there is a variety of defensive measures which the management of a company subject to a hostile takeover bid may explore. Defensive strategies would appear to have reached a higher stage of development in the United States than in Australia. To an extent, this may reflect limitations imposed by Australian company law. For example, Australian companies are currently not permitted by the law to purchase their own shares or, in respect of those listed on main boards, to attach different voting rights to shares. The Australian approach may have helped to avoid some of the excesses that appear to have occurred in the US and may have resulted in a more efficient market in this regard.

The directors of a company have a statutory responsibility to act in the interests of their shareholders; their duty is to advise shareholders of the worth of an offer and to attempt to ensure that the price offered for their shares is the best possible. They may therefore seek to negotiate a higher offer price or provide for a revaluation of assets or revised profit forecasts. Such actions are consistent with the general desirability of ensuring disclosure and the opportunity for shareholders to make an informed response to an offer.

However it is apparent that defensive measures may not always be directed towards the interests of shareholders but may represent efforts to preserve the position of entrenched management. Measures of this kind may be complicated and expensive and may not be designed to facilitate shareholders' decisions or even to allow shareholders actually to consider a bid. Such strategies may include litigation to delay and frustrate bids, share placements and the initiation of cross shareholdings with other targets or potential targets.

It is often suggested that the threat of takeover may encourage companies with low gearing, consistent profitability and a high level of liquid assets to reduce their attractiveness as targets by restructuring or by making substantial acquisitions themselves. However in many cases these 'defensive' actions will result in the company concerned raising its performance and thus its share price - an example of the indirect stimulus to efficiency that the threat of takeovers can induce. In cases where defensive actions do not improve performance, but act to impair it, the company's share price is likely to fall - thereby encouraging takeover offers. Defensive tactics that are not based on improving performance may thus be counterproductive.

One particular concern that has been expressed in this regard is that the threat of takeovers may lead target companies to increase their own gearing through borrowing to undertake takeovers themselves or for other acquisitions. However, the fact that companies with relatively low gearing may be specially attractive as takeover targets can be seen as a market signal that the firms concerned could more profitably utilise their resources in order to give a higher rate of return on shareholders' funds.

It has also been suggested that takeover bids may be affecting the longer-term performance of the economy by concentrating attention on short-term profits. Target companies may reduce investment and other expenditures with longer-term benefits (eg R&D) in order to raise current profits and hence share prices. However, such strategies are likely to be

23

effective only in the very short term and may ultimately be self-defeating. Before too long, the effects of cutting back on investment and longer term planning are likely to become apparent in performance. Such tactics may delay, but are unlikely to prevent, takeovers.

The concern about undue concentration on short term profits also appears to underestimate the sophistication and ability of share market participants in appraising the value and performance of firms. Some overseas studies are relevant in this regard. For example, a recent US Securities and Exchange Commission study examined share market reaction to 62 announcements of R&D projects over 1973 to 1983 and found that the stocks of the companies involved increased by 1.8 per cent (adjusted for movements in the overall market) in the four weeks after the announcement. Another recent US study examined share market responses to announced changes in capital budgets over 1975 to 1981. It was found that, with one industry exception, share prices rose when companies announced increases in capital budgets and fell when companies announced cuts in capital budgets. The exception was the oil industry where the converse applied. However, this appears to have been the result of a stockmarket perception that there was already excess capacity in the US oil industry and that further rationalisation was required(4).

A sizeable proportion of investment in shares is in the hands of institutions such as superannuation funds and insurance companies which need to be concerned with performance over the longer term as well as with the prospects for gain through short term trading. Even if there were systematic differences between the time horizons of such investors and those of company managers, it is by no means obvious that those of managers are more likely to coincide with the national interest. (There are many examples of firms which have failed through their managers' preoccupations with long-term projects that never reached commercial success.) Takeovers may be seen as a means of encouraging management to act more in line with shareholder preferences. While there may be reasons for the community to prefer a higher level of investment than would be set by shareholder preference, discouraging takeovers would be an indirect and inefficient way to achieve this.

4.3 PARTIAL BIDS

Many of the potential benefits of a takeover to a bidder derive from obtaining control of the target company rather than full ownership. Due to the dispersion of share ownership of many companies, control or dominant influence can frequently be obtained with substantially less than 50 per cent ownership. Many recent takeovers in Australia have been partial bids. Companies legislation has attempted to regulate such offers with the objective of providing fairness of treatment between shareholders in the target company and to discourage practices which may have a coercive effect in encouraging early selling.

----------

(4) See A. Ehrbar, 'Have US Takeovers Gone Too Far', Fortune, May 27 1985, pp. 14-18 and Council of Economic Advisers, 'Annual Report' in Economic Report of the President 1985, (Washington: US Government Printing Office, 1985), pp. 201-2.

24

Particular concerns include that:

• all shareholders should have the opportunity to sell their shares on equal terms;

• shareholders may accept an offer not on its merits but under coercion for fear that they will become a locked-in minority (for example if they fear that minorities will be treated less well by the new majority shareholders than by the old); and

• the risks in a takeover situation should fall on the bidder rather than the shareholders of the target company.

In December 1985 the Ministerial Council for Companies and Securities announced measures to regulate partial offers which the Commonwealth and State Governments are required to implement in accordance with the Agreement for a cooperative companies and securities scheme. A bill incorporating the following measures has now passed through the House of Representatives:

• the limiting of partial bids to a proportional basis;

• prohibition of maximum acceptance conditions in partial bids;

• escalation clauses entered into by a bidder within six months of a takeover bid may not be triggered by higher prices offered under the bid;

• all takeover offers to reflect in any cash bid the highest pre-bid price in the previous four months; and

• provision for a company to impose a requirement for a shareholder plebiscite and majority acceptance before a partial bid may proceed.

The objective of the legislation is to promote equity amongst target company shareholders by attempting to reduce the uncertainties faced by them in the context of partial bids. At the same time the proposed arrangement may increase the uncertainty and the costs borne by those seeking to make partial offers. It remains to be seen whether the changes which will be introduced by the proposed legislation will have a significant effect on the overall level of takeover activity.

The public debate surrounding the regulation of partial takeovers has raised a number of issues in addition to the concerns referred to above. For example, it has been argued that:

• there is an underlying conflict in such regulation between the objective of fair treatment for shareholders and the reality that shareholdings of different sizes may be of differing strategic importance for control of the target company and thus of differing market values;

• the coercive pressure in partial takeover situations would appear to be related more to the partial nature of the bid rather than its form (pro-rata or proportional); shareholders will be conscious that the share price may have been inflated by the bid and that when the offer is concluded, the share price may fall, possibly to below its pre-announcement level; and

25

• even 'locked-in' shareholders could gain in the longer-run if company performance (and hence the share price) is improved due to the takeover. Empirical evidence from the US and Australia is quoted as suggesting that post-takeover prices do remain above their initial, pre-announcement level, bringing into question the need for extensive protection of shareholders in target firms.

More fundamentally, concern has increasingly been expressed as to whether the Takeovers Code achieves the objectives of ensuring that the acquisition of shares in companies takes place in an efficient, informed and competitive market. There have been calls for a review of the Companies (Acquisition of Shares) Act, both in relation to its substance and overall complexity. Relevant issues include the appropriateness of the level of disclosure of substantial shareholders in a company, the mechanism of disclosing the identity of shareholders (section 261 notice) and the scope which may exist for excessive litigation. The Ministerial Council has indicated that, in the light of recent developments in the share market, it will be conducting a general review of takeovers at its meeting next July.

5. CONCLUSION

Not all takeovers necessarily increase economic efficiency. Some may be misjudged; all can consume substantial amounts of managerial time and resources. But government 'vetting' mechanisms are unlikely to be able to distinguish between takeovers that will effectively improve returns from existing assets and those that will not. That particular function is best left to the market itself.

These considerations suggest that policy should neither actively encourage nor discourage takeover activity, ie it should be neutral in its impact. Policy should be directed towards fostering efficient and informed share markets that facilitate the monitoring of company performance but not provide either undue encouragement or discouragement for takeovers. There may be scope for regulatory changes in pursuit of that objective.

Where there are particular reasons for concerns about the effects of takeovers - for example in restricting competition, in giving rise to costs to tax revenue, or in relation to foreign ownership and control - the appropriate course is to make adjustments to the policies applying to those particular areas, rather than action directed to restrict takeovers as such. In the case of taxation concerns, the introduction of the imputation system for company taxation in July 1987 and the comprehensive capital gains tax now in place should largely remove any present bias towards debt over equity financing which may have been relevant to some takeovers. Taxation measures designed to restrict takeover activity would have no sound basis in economic principle, would be readily circumvented and could do serious damage to the operation of equity markets in the short period prior to the commencement of the imputation system.

Concerns about the impact of overseas borrowings to finance takeovers on the balance of payments and the volume of overseas debt appear to be reflections of the •domestic concerns•. That is, takeover activity as such does not raise additional issues from a specifically external account perspective.

26

ATTACHMENT A -THE COST TO REVENUE OF TAKEOVER ACTIVITY

Takeover involves a change of ownership of shares. The means of acquiring the target shares might be provided by:

• borrowing, whether from home or abroad;

• activation of idle cash balances;

• liquidation of physical assets which may be 'income-producing'; or

• a share-swap or new share issue, whether directly or indirectly, possibly involving one or more third parties.

The tax revenue implications of the share acquisition will depend upon which of these is exercised, as well as a host of other factors. In general, the takeover activity could have indirect effects on asset rates of return, and possibly on output and employment, each of which would have implications for aggregate tax revenue.

To keep the investigation tractable, attention focuses here on a snapshot comparison of tax revenue collected from various agents before and after the takeover, ignoring any of the possible indirect effects mentioned above. A simple illustrative case is presented and then some further issues discussed. While the first case may appear to contain some complexities, it is highly simplified by real world standards.

A Simple Illustrative Case

In this first case, the takeover is financed partly by equity and partly by borrowing. For the sake of simplicity, it is assumed that the borrowing is directly from those shareholders of the target company who sell and that equity is also raised from those shareholders by way of a rights issue (this apparently artificial assumption simplifies the issues but is relaxed in the analysis below). The shareholders acquire a debt instrument and some new equity in exchange for pure equity, and conversely for the share purchaser. If the interest bill were fully tax deductible (which requires that the purchaser have sufficient taxable income to absorb it), and the borrower and lender had the same tax rates, there would be no cost to revenue.

For example, suppose $600 million is borrowed at an interest rate of 15 per cent per annum, a rights issue of $400 million is made, and that dividends are $50 million per annum. Suppose tax rates are 46 per cent for each party. Shareholders exchange $1,000 million value of shares for $600 million cash, which is lent to the share purchaser and $400 million of rights issued by the share purchaser. The income and income tax revenue implications are as shown in the following table.

27

 

Pre-takeover

Post-takeover

Change

Agent

Income

Tax

Income

Tax

Income

Tax

             
     

($ million per annum)

   

Share Buyer

   

-40 (a)

-18.4

-40

-18.4

Share Seller

50

23

90

41.4

40

18.4

Totals

50

23

50

23

0

0

(a) $40 = $50 (dividends) -$90 (interest; .15x$600)

As a first complication, it should be recognised that the composition of shareholders as between tax exempt, non-resident, and fully taxable resident categories matters. Suppose the composition of shareholders who sell is as follows:

Shareholder

Per Cent of Shareholders

Type

Resident

Total

Resident

 
 

1. Taxable

65.2

48.9

2. Exempt

34.8

26.1

 

100.0

75.0

Non-resident

NA

25.0

Total

 

100.0

Suppose all of the rights are accepted by domestic resident shareholders (the 'bottom line' is no different if this is not the case, but the analysis is somewhat more complicated). Since domestic shareholders decide to accept $400 million in rights but sell (.75x$1,000=) $750 million value of shares, on the foregoing assumption they must lend the other ($750-$400=) $350 million which they receive. Non-resident shareholders lend the balance of $250 million borrowed by the share purchaser (ie their total proceeds). Hence, of a total interest bill of (.15x$600=) $90 million, (.15x$350=) $52.5 million is received by residents, and (.15x$250=) $37.5 million by non-residents.

The share purchaser will usually be a company, receiving a section 46 dividend rebate (that is, dividends will not be taxed in the hands of the company). There will still be tax on the dividends to the extent they are distributed and taxed at 46 per cent, which is assumed to be the case in the foregoing table.

28

In the example considered here the (corporate) purchaser's cash flow falls by $40 million per annum before tax as a result of the takeover and by $21.6 million after tax. Hence, it is possible that none of the additional dividend income would be distributed to that company's shareholders. In that event, if sections 46 and 50 do not operate to require interest to be offset first against rebatable dividends, there is then a 'cost to revenue' of $23 million per annum. This cost arises purely because of dividend retention which in turn is only a significant cost to revenue under the current classical system of company tax. Under a system of full imputation of company tax dividends distributed to resident shareholders are essentially not taxed in their hands, as the imputation credit frees taxable shareholders from tax on their dividends. Additional retentions would not involve any loss of personal tax revenue from resident shareholders such as occurs under the classical system. (There could still be some loss of dividend withholding tax if dividends paid to non-residents were reduced or if the retention reduced excess compensatory tax on distributions out of untaxed income. There would be offsetting gains to revenue in cases where imputation credits on distributions would have exceeded the shareholders' personal marginal rates and the excess would have offset tax on their non-dividend income.) Nevertheless in the present classical taxation system an increase in aggregate dividend retention is the first possible source of a cost to revenue. Note that in the present illustration, there would be the same implications for revenue if, without specifically focusing on the additional dividend income, for some reason the target company had simply decided to retain additional earnings of $50 million per year.

In all of what follows it will be assumed that no additional dividends acquired by the purchasing company are distributed to its shareholders. This assumption is responsible for $23 million per annum of any cost to revenue subsequently computed (or less than $23 million to the extent the marginal rate of ultimate shareholders is less than 46 per cent).

Borrowing and lending interest rates of financial intermediaries will not usually be equal. Suppose the rate they offer to lenders is 13 per cent (the rate they charge borrowers remains at 15 per cent). Furthermore, tax rates will not be the same. Suppose the composition of tax rates is as follows:

 

Tax Rate on

Tax Rate on

Shareholder

Dividends

Interest

Type

 
 

Resident

 
 

Taxable

0.46

0.46

Exempt

0.00

0.00

 
 
 

Weighted Average (c)

0.30

0.30

 
 
 

Non-resident

0.15(a)

0.10(b)

Weighted Average Tax Rates (c):

0.2625

0.25

(a) Dividend withholding tax rate for treaty countries.

 
 

(b) Interest withholding tax rate

 
 

(c) Weights as per table above.

 
 

29

The margin between borrowing and lending rates goes to banks. In total, they receive ((.15-.l3)x$600=) $12 million. Suppose 60 per cent goes to resident banks with a tax rate of 46 per cent, the remainder to non-residents overseas.

The other complication to be confronted is that the share purchaser will not usually have unlimited residual taxable income against which to set-off all of the interest bill of $90 million. Suppose the purchase of the shares is marginal in a cash flow sense. That is, suppose the share purchaser has just enough retained earnings to meet the net drain on cash flow associated with the share purchase. The following table presents an example of such a marginal purchase:

 

Cash Flow

 
 

Pre-Takeover

Post-Takeover

 

($ million)

 
 
 
 

Gross Income

100

150 (b)

 
 
 

Expenses

50

140 (c)

 
 
 

Effective Taxable

 
 
 
 
 

Income

50 (a)

0 (d)

 
 
 

Tax

23

0

 
 
 

Post-Tax Income

27

10 (e)

 
 
 

Distributions

10

10

 
 
 

Retained Earnings

17

0

(a) All expenses deductible.

(b) $150 = $100 + $50 (dividends).

(c) $140 = $50 + $90 (interest).

(d) s.46 rebate means dividends not taxed. Additional interest of $90 million more than completely off-sets residual taxable income of $50 million. (There is 'excess interest, of ($90-$50=) $40 million).

(e) $10 = $150 - $140.

The investment illustrated in the table is marginal in the sense that the net drain on cash flow runs down retained earnings to zero. In this case, tax paid by the share purchaser falls by $23 million.

30

The income and income tax revenue implications of the takeover for the share buyers and sellers and banks are:

 

Pre-takeover

Post-takeover

Change

Agent

Income

Tax

Income

Tax

Income

Tax

 
 
 

($ million)

 
 
 
 
 
 
 
 
 

Share Buyer

50(b)

23

10(c)

0

-40

-23

Share Seller

50

13.125(a)

78(d)

16.9(e)

28

3.775

 
 
 
 
 
 
 

Banks

 
 
 
 
 
 

1.Domestic

 
 

7.2

3.3(f)

7.2

3.3

2.Overseas

 
 

4.8

0

4.8

0

Totals

100

36.125

100

20.2

0

-15.925

(a) $50 x .2625 = $13.125.

(b) $50 = $100 (Gross Income) - $50 (Expenses).

(c) $10 = $150 $140.

(d) $600 x .13 $78.

(e) (.13x$350=) $45.5 x .3 + (.13x$250=) $32.5 x .1 = $16.9.

(f) $3.3 = ($12x.6=) $7.2 x .46.

The net cost to revenue is $15.925 million. In reaching that figure it has been assumed that the new issue of rights did not lead to any increase in total dividend pay-out by the company which made the issue. Had there been any increase, there would have been more tax on the shareholders and a smaller cost to revenue.

Other Issues

If sections 46 and 50 were made to operate so that interest offset first against rebatable dividends the cost to revenue would, in the case illustrated here, fall by $4.6 million. The reason is that net interest of only ($90-$50=) $40 million could be offset against the residual taxable income of $50 million, leaving $10 million still taxable at 46 per cent.

31

The preceding analysis used what would normally only be a convenient fiction, that it is the shareholders who sell who also accept rights and lend to the share purchaser. In fact, it makes no difference to the estimated cost to revenue if loans are raised elsewhere, but may affect the cost if rights are purchased elsewhere. For example, suppose (in the simplest case) the share purchaser borrows all $600 million overseas, but shareholders who receive cash of $1,000 million invest all but $400 million used to purchase rights in the domestic financial market. Then the shareholder injection of $600 million will simply replace some other overseas borrowing of $600 million, and there would be no net tax implications. However, if all $400 million rights were purchased by non-residents, it can be shown that the cost to revenue would fall by (.13x$400x(.3-.1)=) $10.4 million.

Suppose there were no rights issue, and all $1,000 million were borrowed. Domestic resident shareholders would receive cash of $750 million, and non-resident shareholders would receive $250 million. This would not affect the value of the interest deduction to the share purchaser; his tax would still fall by $46 million, 'excess interest' increasing from $40 million to $100 million. However, share sellers would obtain income of (.13x$1,000=) $130 million, on which tax of (.3x$97.5+.1x$32.5=) $32.5 million would be collected, an increase of ($32.5-$16.9=) $15.6 million. Tax revenue from banks would increase by $2.2 million because domestic (resident) banks receive an additional $4.8 million income. The net result is that the cost to revenue falls by ($15.6+$2.2=) $17.8 million. In the present illustration, there would therefore actually be a 'gain to revenue' of ($17.8-$15.925=) $1.875 million.

In all of the above, the cost to revenue of a takeover has been calculated by comparing snapshots of revenue collected before the takeover and revenue collected after the takeover. This is not an appropriate comparison if the particular takeover replaces or 'crowds-out' some other activity which would also have involved a cost to revenue. It is inconceivable that there would be no such effect. The appropriate method would then be to compare the estimated cost to revenue of the takeover with the cost to revenue of the activity which is crowded-out, the net cost to revenue being the difference between the two.

Some of the implications of such 'crowding-out' are illustrated in the following example. Suppose that in the 'simple case' numerical illustration presented above, the purchase of the $1,000 million value of shares had been financed by the sale of $1,000 million of income-producing assets in which the seller had no equity (that is, the assets were purchased with borrowed funds). Suppose the rate of interest on the borrowed funds is 15 per cent, but that the rate of return on the physical assets is only 5 per cent. The agent effectively exchanges shares earning the same nominal yield ($50 $1,000=.05) for the physical assets. In this case there is no net cost to revenue. In each case, tax is lower than what it would be in the absence of either investment by (.46x(.15x$l000-$50)=) $46 million per annum. Similar arguments apply in each of the more complicated scenarios presented above, the only possible source of a cost to revenue being dividend retention.

32

Takeover activity financed by borrowing is often simply a means of increasing gearing, or more often, re-distributing gearing among agents. In general, increasing or re-distributing gearing will involve a 'cost to revenue' if borrower and lender tax rates differ. In this sense there is nothing special about the nominal cost to revenue of a takeover per se. Furthermore, it seems likely that in many cases the takeover substitutes for (and therefore 'crowds-out') some other means of increasing gearing. The net cost to revenue of a takeover per se is in such cases likely to be minimal.

To summarize, there is nominally an aggregate cost to revenue associated with takeover activity if, but only if, as a result of that activity:

• the aggregate (economy-wide) dividend pay-out ratio is reduced; and/or

• aggregate gearing is increased; and/or

• the distribution of aggregate gearing among agents is affected, and

• there is sufficient borrower taxable income against which to offset all additional interest, and

• borrower and lender tax rates differ (the former exceeding the latter), and/or

• there is a margin between borrowing and lending interest rates which is not taxed at the same rate at which additional interest is deductible.

None of these factors is peculiar to takeover activity; ie they need not arise as a result of such activity, and they may arise as a result of other (non-takeover) activity.

Finally, as noted above, under a system of full imputation of company tax the revenue implications of any takeover-induced effect on dividend retention would be slight.

33

ATTACHMENT B - SOME DATA ON GEARING

Debt/equity (D/E) ratios for Australia are compiled by the Reserve Bank (RBA) and the Sydney Stock Exchange.

The Stock Exchange ratios are based on a somewhat narrower definition of debt than those of the RBA, being calculated by dividing all financial debt by ordinary equity, net of intangibles. The RBA debt measure is equal to total liabilities less shareholders' funds and minority interests, as a proportion of shareholders' funds and minority interests. The RBA measure therefore includes not only financial debt, but other forms of debt such as provisions for dividends, taxation, superannuation and long service leave. Both the RBA and the Stock Exchange D/E ratios measure equity by its book value.

The RBA's D/E ratio covers the whole corporate sector whereas the Stock Exchange measure covers only companies listed on the Sydney Stock Exchange.

The RBA data for Australia for the period from 1960 to 1984 are set out in Table 1 while the Stock Exchange data back to 1976-77 are shown in Table 2.

The RBA data indicate a secular upward trend in the D/E ratio over the period. The Sydney Stock Exchange data (over a much shorter period) show no consistent trend.

It may be noted that the higher levels of inflation experienced in the seventies and eighties is likely to have been a factor in the increase in D/E ratios during that period compared with the sixties:

• on the equity side, the reliance on book values is likely significantly to understate the current value of assets (and therefore equity) during periods of high inflation; the March 1985 Company Finance Supplement to the Reserve Bank Bulletin notes that 'the aggregation of data from published accounts cannot fully reflect the impact of price movements'; and

• on the debt side, the rising cost of labour and capital during periods of high inflation is likely to have required concomitant increases in debt; additionally, the low (and at times negative) real rates of interest experienced in the seventies and early eighties provided a strong incentive to the use of debt as a financing instrument.

The extent of the contribution of this measurement factor to the upward trend in the D/E ratio is not known. To the extent that assets are now revalued more frequently than in the past, the inflation bias would be limited though not removed. It is probably the case that the trend evident in Table I partly reflects more fundamental factors.

34

TABLE 1: RBA D/E RATIO

 

D/E RATIO

   
       

YEAR (a)

ALL INDUSTRIALS

MINING

TOTAL

     

NON-FINANCIAL

1960

0.68

na

na

1961

0.64

na

na

1962

0.63

na

na

1963

0.65

na

na

1964

0.68

na

na

1965

0.70

na

na

1966

0.70

na

na

1967

0.70

na

na

1968

0.71

na

na

1969

0.75

na

na

1970

0.78

na

na

1971

0.82

na

na

1972

0.83

na

na

1973

0.84

na

na

1974

0.98

na

na

1975

0.97

na

na

1976

1.00

0.95

1.00

1977

1.03

0.91

1.01

1978

1.03

0.84

1.00

1979

1.06

0.84

1.03

1980

1.07

0.66

1.00

1981

1.06

0.64

0.98

1982

1.13

0.78

1.06

1983

1.11

0.89

1.07

1984

1.17(b)

1.09(b)

1.16(b)

(a) Based on data reported by companies covered by the survey within the calendar year.

(b) Preliminary estimate, subject to revision.

35

TABLE 2: SYDNEY STOCK EXCHANGE D E RATIO

Year (a)(b)

D/E

 

RATIO

1976/77

0.47

1977/78

0.46

1978/79

0.45

1979/80

0.43

1980/81

0.38

1981/82

0.39

1982/83

0.47

1983/84

0.57

1984/85

0.53

(a) Data for earlier years (back to 1974/5) are available but not on a consistent basis.

(b) Based on the income tax year of companies covered by the Stock Exchange study.

A recent article in the Morgan Stanley 'Economic Perspectives' entitled 'Corporate Restructuring, 'Junk' and Leverage: Too Much or Too Little' contained D/E ratios for the US, Japan and West Germany for 1984, these ratios are set out below:

 

COUNTRY

D/E RATIOS

D/E RATIOS

   

(BOOK EQUITY VALUES)

(MARKET EQUITY VALUES)

       

(i)

US

1.22

1.25

(ii)

Japan

2.33

1.14

(iii)

West Germany

3.63

1.57

From these it can be seen that the RBA's Australian D/E ratio (based on the book value of equity) of 1.16 is well below the D/E ratios (based on book values) reported for Japan and West Germany and comparable to that of the US. Caution should, however, be used in placing undue weight on such comparisons as it is not clear whether comparable definitions have been used in deriving the D/E ratios in the Morgan Stanley paper and those published by the RBA (which are, however, based on a broad definition of debt).

Differences in the scope and coverage of the measures also present problems in interpretation. The RBA measure is derived for the corporate sector as a whole using a broad sample of data obtained for both listed and unlisted companies, whereas the Morgan Stanley measures are derived by averaging data for the five largest corporations in each of seven key industries in each country.

36

The Morgan Stanley article also includes a D/E measure based on the market value of equity which allows the rates of return (ie profitability) of a company to be taken into account in assessing its ability to carry and service debt. No such measure is published for Australia by either the RBA or the Sydney Stock Exchange.

37

ATTACHMENT C - THE SECTION 46 DIVIDEND REBATE

The operation of sections 46 and 50 has been raised in the context of takeover and redeemable preference share activities. Independently of their relevance to such activities, there is a need to examine whether those sections currently have the effect they should, and whether amendment of the law is desirable.

This Attachment considers questions of principle, of case law and past legislation, and of administrative feasibility.

Appendix C1 describes briefly the legislated provisions of sections 46 and 50. The wording of sub-section 46(7) provides for any expenses incurred in order to earn dividend income to be deducted from the gross dividends and for the rebate to be be based on the net amount. The magnitude of the rebate therefore depends importantly on the scope of any expenses deducted. However, neither section 46 nor section 50 defines with any specificity what, if any, expenses are in that category. They merely provide for the possibility that there may be some. It has been left to case law to determine scope.

Appendix C2 describes the history of section 46 and some key developments in case law and legislation on the scope of deductions, as well as some possible amendments considered, but not proceeded with, in 1971. The present position is that, except where the dividends are received by life insurance companies, only expenses (including interest) incurred exclusively in connection with the earning of dividends need be offset against dividend income to arrive at dividends included in taxable income, on which the section 46 rebate is based.

That raises a question as to how, in principle, the rebate should work. Specifically, what expenses should be offset against gross dividends to arrive at the amount eligible for rebate? The question has come up before, and is not straightforward. Given the way the present rebate developed before 1940 as part of a common dual-purpose rebate applying both to companies and to individuals - it would be a mistake to read too much regarding the original intention for treatment of companies into the wording of sections 46 and 50 in the form in which they have come down to us. Certainly the possibility of there being deductions against dividends is allowed for, but, given the origin of the provisions, there is no ground to read into their wording a definite intention that any particular type of deduction should be made against dividends or that there be an apportionment of expenses.

That question should now be considered afresh as a question of principle, practicality, and economic effect, having specific regard to the only remaining role of section 46, now that it no longer applies to individuals.

In order to clarify the fundamental purpose of section 46, it is helpful to consider first how dividends should appropriately be taxed. Appendix C3 considers that issue. It is concluded there that the full imputation system will provide appropriate taxation of dividends, by serving (in general) to tax dividends effectively only once, at the marginal tax rate applicable to the individual dividend recipient. This outcome would also be secured under imputation where dividends pass through successive layers of companies, if the imputation credit were allowed to roll through these layers as the method of freeing successive companies from tax on the dividends.

38

If, however, the section 46 rebate is retained, it is further shown in Appendix C3 that in order to obtain the same (appropriate) outcome as under rolling imputation, it is necessary that the rebate be calculated on the gross dividends received by a company.

What imputation aims to do is to ensure that the dividend is effectively taxed only once, regardless of the number of company tax layers it is passing through. Under imputation, section 46 can be seen as seeking to achieve the same result.

Even in the context of the classical system, section 46 can be seen as seeking to ensure this treatment so far as the interposition of further companies is concerned, after the dividend has first borne tax at the level of the original company generating it.

However, the classical system of company taxation lacks the economic logic of an imputation system and, on conceptual grounds, it is much more difficult to come to a conclusion as to how the operation of section 46 best meets the objectives of the classical system. It is concluded in Appendix C3 that, on balance, a gross rebate remains appropriate but the conclusion is more problematical.

In the case of associated companies under the classical system, however, it is unambiguous that the rebate should be on a gross basis.

Moreover, on pragmatic grounds, the case for a change in the status quo is anything but strong:

• the current gross basis has effectively been in operation for the bulk of the taxpaying population for somewhere between 30 and 55 years (see Appendix C2); and

• the gross basis will be the more appropriate basis under the imputation system to be introduced in a little over twelve months time.

If, despite the foregoing, the treatment were to be changed for other than associated company or intra-group dividends, the practical difficulties identified in 1971 and discussed in Appendix C2 would again arise. They include:

• the extent of common ownership to qualify for associated company treatment. The company arrangements in place which could be affected go well beyond the 100 per cent ownership test for the group loss transfer provisions, but to extend associated company treatment to a lesser degree of ownership in the case of section 46 would put strains on the maintenance of the 100 per cent test in the group loss provisions;

• how to allocate expenses to different categories of income. To attempt to identify the purpose of each and every expenditure item as a basis for allocation would be virtually impossible, as there would be a large proportion of expenditure which clearly served multiple purposes. Arbitrary formulae would have to do some or all of the job; and

• in order to ensure that other companies could not take advantage of the associated company outlet, complications would have to be built into the legislation. If all else failed, companies determined to borrow heavily, say, for takeover purposes, could set up a company structure in which one company borrowed and a different one acquired shares. To defeat that would require very complicated tracing rules and other provisions; the feasibility of administering them, and their effectiveness, would be open to question.

39

This last point gets to the crux of the issue. Money is, of course, fungible. It is, in terms of economic logic, impossible to allocate particular slices of a firm's total funds employed to particular aspects of its business. Moreover, as a practical matter, attempts to apply tracing rules can almost universally be avoided by minor rearrangements of financing transactions designed to prevent an investment on which interest deductibility would be denied from being associated with the debt incurred to finance its acquisition.

The fundamental issue of fungibility would be likely to render any amendment to section 46 largely ineffective, including in respect of takeovers. There would also be a question, in the case of successful takeovers, whether any restraining effect would be more than temporary. Unlike the takeover moves that have recently been publicized, many takeovers seek a majority or all of the shares in the target company. Once they succeed, the way to associated company treatment may be open to them, or they could merge and there would be no dividend for the interest to offset.

In takeover situations there is much more at stake than the dividends which might flow from the target company. The dividends would be only a small element of the takeover strategy if the interest paid by the raider greatly exceeded the dividend income. The bulk of the interest would be deductible unless very arbitrary rules were applied. The effect of amending section 46 could be minimal (or, as already noted, of no effect at all if the companies merged).

Whenever considering the case for legislation, it is useful to distinguish situations where

the measure would remove the cause of a problem from situations where it could penalize the consequence and therefore have a restraining influence on the scale of the problem. An example of the latter was the use of redeemable preference shares. A change to the allocation rules for section 46 purposes of interest costs of banks could have discouraged the latter's take up of such shares, but the real cause of that problem lay elsewhere - the different tax treatment of interest and preference dividends for the 'borrower'. The measures recently announced attacked the latter directly.

Any judgment concerning whether to change the existing treatment of the section 46 rebate is necessarily an on balance one:

• conceptual considerations are clear cut in the case of imputation and of associated companies within either the classical or the imputation system. They are more ambiguous in the case of unassociated companies within the classical system. Overall, the balance of conceptual argument favours a gross treatment but is not sufficient, in itself, to conclude the matter;

• pragmatic considerations - most particularly the fungibility of money, but also the historical experience and the pending introduction of imputation - argue strongly for no change; and

• legislative intention considerations are, in this case, largely irrelevant, given that the provision which has come down to us is merely what remains of a pre-war provision which had a dual purpose.

The balance of considerations supports the status quo.

40

APPENDIX C1 - SECTIONS 46 AND SO: LEGISLATION

(a) Section 46

Gross dividends received by a resident company are assessable income. They are added to other assessable income, and allowable deductions are deducted to arrive at the company's taxable income. The company tax rate is applied to taxable income to obtain the tax assessed on it.

If that were the end of the matter, income already taxed in the hands of the first company which earned it would be taxed again as dividend income in the hands of each company interposed between the first company and the ultimate individual shareholders. Section 46 prevents such accumulation of further tax as dividends pass from company to company before finally leaving the company sector; it allows a rebate of tax on dividends included in taxable income of companies.

For example, if a company had assessable income comprising $100 of dividends and $1000 of other income, and allowable deductions of $800, its taxable income would be $300, and tax assessed on that amount at the present rate of 46 per cent would be $138. That tax would be reduced by a rebate, at the company's rate of 46 per cent, on the amount of dividends included in its taxable income.

Sub-section 46(7) provides that 'deductions allowable to the shareholder under this Act from income from dividends' are deducted from gross dividends included in assessable income to arrive at 'dividends included in taxable income'. In the foregoing example, the rebate would be calculated on the $100 of gross dividends less that part of the allowable deductions that relate to the earning of the dividends. Consequently:

• if none of the deductions related to the earning of dividends, a rebate would be allowed on the entire $100 of dividends. The rebate would be $46 and tax payable would be reduced to $92; or

• if some of the deductions related to the earning of dividends, the rebate would be allowed on less than $100. For example, if $20 of deductions related to dividends, there would be $80 of dividends 'included in taxable income', the rebate would be $36.80, and tax payable would be reduced to $101.20.

(b) Section 50

Section 50 sets out rules which are operative where it is necessary to break taxable income into the components which are, respectively, dividends, other property income and personal exertion income (the latter including, in the case of companies, income from business activities), and the deductions incurred relate to one or more of those

41

components. Where the deductions exceed the corresponding component of gross assessable income against which they are first deductible, it then becomes necessary to specify which of the other components will first be reduced by deduction of the excess to determine the break-up of taxable income.

In the case of public companies, section 50 in conjunction with section 46 provides:

(i) to the extent that expenses relate to dividend income they are deductible first against that income and, if unabsorbed, then successively against other property income and personal exertion income;

(ii) where expenses relate directly to other property income they are deductible first against that income then successively against dividend income and personal exertion income; and

(iii) in any other case, expenses are deductible first against personal exertion income, then successively against other property income and dividend income.

(In the case of a private company, the relevant provisions go further and, in effect, require the Commissioner of Taxation to apportion interest on multi-purpose loans (or other multi-purpose expense) indirectly related to the receipt of private company dividends.)

As is explained in Appendix C2, section 50 had a more significant role before 1940 when the dividend rebate applied to individuals as well as companies and the rates of personal tax on property income were higher than those on personal exertion income. The rebate ceased to apply to individuals in 1940 and the differential rate was removed in 1952 and, apart from a property income surcharge which applied only in the 1974-75 income year, has not been in the system since then.

These days, section 50 is largely a vestigial provision, although circumstances can arise where, in its present form, it would have some effect. An instance arose in the Palvestments case in 1965, referred to in Appendix C2. And a few years ago, the Auditor General drew attention to some transactions of the AIDC which were undertaken to prevent an outcome where some of its deductions would have been set against its dividend income and, by reduction of the section 46 rebate, the tax savings from them lost.

Viewed in that light, some commentary on takeovers can be seen to have given section 50 more prominence than it deserves, instead of the main provision for deduction of expenses relating to dividends against dividend income which is contained in sub-section 46(7). The interpretation of that provision, over the years, is set out in Appendix C2 and the principles surrounding the issue are canvassed in Appendix C3.

42

APPENDIX C2 - SECTION 46: DEVELOPMENTS IN CASE LAW AND LEGISLATION

Pre-War Arrangements

Prior to World War II, the predecessor of section 46 in the Commonwealth income tax law applied to both individuals and companies. It allowed both categories of taxpayer a rebate on 'dividends included in taxable income', calculated at the lesser of the taxpayer's average rate of tax on his taxable income or the rate of tax borne on the income by the company which paid the dividend.

At that time, therefore, the one rebate provision had a dual purpose. In its application to companies, it had the same purpose as now - to ensure that, after income had borne company tax once, there were not further amounts of company tax on it as it passed as dividends from company to company before reaching individual shareholders. In its application to individuals, its purpose was to prevent the 'double taxation of dividends' which now occurs under the present classical system of company tax. In 1940, as a wartime revenue-raising measure, the rebate ceased to apply to dividends received by individuals; that marked the beginning of the classical system in Australia. Since then the rebate has applied only to companies.

It is possible that the original design of the single measure to serve two purposes may have involved a degree of compromise between the two objectives. In the case of companies the relevant point was the effect on dividends passing through to the next stage, whereas in the case of individuals there was no next stage. Certainly the order in which allowable deductions were offset against the different components of assessable income (dividends, other property income, and personal exertion income) was then a more significant matter for individuals than for companies. That was partly because the personal tax rates on property income (including dividends) were then higher than those on personal exertion income. Also concessional allowances for dependents at that time took the form of deductions. The order in which they offset different components of assessable income would have affected the average rate of tax on the taxpayer's taxable income. That would also have been true of personal allowances which freed a first slice of income from tax.

At that time company tax collected by the Commonwealth was at a rate of 5 per cent. In so far as there may have been any compromise in the design of the dual purpose rebate, which made it less than ideal in its application to companies, that would have been a much less serious matter than it is now.

43

Case Law

Section 46 provides that a resident company is entitled to a rebate, at the average rate of tax on its taxable income, on dividends 'included in its taxable income'. However, taxable income is a composite figure arrived at by subtracting all of a taxpayer's allowable deductions from all of its assessable (gross) income. Sub-section 46(7) provides that 'deductions allowable to the shareholder under this Act from income from dividends' are to be deducted from dividends included in assessable income to arrive at the amount of dividends included in taxable income.

However, the section is somewhat ambiguous as to the precise manner of determining the deductions from dividends, and consequently as to the precise amount of rebate. Broadly speaking, there could be two approaches to the question:

• apportion the taxpayer's allowable deductions and set off a fair proportion of them against gross dividend income to determine dividends included in taxable income; or

• set off only those deductions which would not have been incurred at all if the dividends had not been received.

In a series of decisions between 1931 and 1953 the High Court tended to apply the latter approach. It gradually emerged that virtually no deductions would be subtracted from dividends in calculating the rebate unless they related exclusively to the earning of dividends. (The only exception is that the amount subject to rebate cannot exceed the company's taxable income.) This approach also emerged in a 1953 decision, on an appeal by the AMP Society, which carried the implication that no part of a life office's section 113 deduction (for allowable management expenses) or section 115 deduction (for a percentage of calculated liabilities) could be taken into account in calculating the rebate. The rebate had to be calculated on the gross amount of dividends. The life offices were the principal beneficiaries of this treatment of expenses.

In August 1971 the foregoing trend reached a point, in the High Court's decision in the Investment and Merchant Finance Corporation case, where the effectiveness of classic dividend stripping arrangements was established as a form of tax avoidance. It relates to a situation where a moribund company has substantial assets representing undistributed profits which, if distributed as dividends, would be taxed in the shareholder's hands. Instead of proceeding that way, the company's shares were sold to a share dealing company for a price corresponding to their asset backing, so that the former shareholders received that amount as tax-free capital gains. The typical steps for the share dealing company to take then were:

• have distributed to itself as a dividend the undistributed profits of the moribund company, which left the shares of the latter without any asset backing and therefore worthless;

44

• in pre-tax terms, the dividend received by the share dealing company would be offset by its loss on disposal of the now worthless shares; and

• but in post-tax terms the two were not offset against each other. Rather the gross dividend was freed from tax by the section 46 rebate, and the loss on the shares was set against the share dealing company's income from other sources. The outcome effectively amounted to a double exemption for the dividends.

Amending Legislation

After consideration of the implications of the decision by the Government of the day, the then Treasurer issued a press release on 31 August 1971 indicating the intention to amend the law not only so as to deal with dividend stripping, but also more generally to ensure that a share of all allowable deductions should be set against gross dividends in determining the amount of section 46 rebate.

Immediately following that announcement, representations were received from a number of quarters concerning the proposed rationalisation of the allocation of deductions between rebatable dividends and other income. The representations supported the need for legislation to counter dividend stripping and other tax avoidance schemes based on the section 46 rebate, but opposed any move which might have resulted in part of company profits passing through a number of companies before distribution to taxable shareholders being taxed more than once in the hands of the companies.

For example, one company structure was such that the function of the parent company was to supply both general and financial management services to its wholly-owned operating subsidiaries. It charged the subsidiaries a management fee for the services, which covered the costs of providing them. The costs to the parent of handling dividends received from the subsidiaries or on other investments were negligible. If the subsidiaries had done the servicing work themselves rather than pay the parent to do it, there would have been no risk of the costs reducing the parent's section 46 rebate. Had it not been for the long-standing basis on which section 46 had been interpreted, the company may have arranged the activities performed by different parts of its structure in an alternative manner. But their structure had been put in place to fit the law as they knew it, and they considered it would be unreasonable for the proposed changes to cut back their rebate.

The representations demonstrated that, if the law were amended along the lines then proposed, some untoward double taxation could occur under established corporate structures in the transfer of profits by way of dividends passing between companies in the one group. While these results could have been avoided by re-organisation of affected groups, the companies naturally did not want to go to that trouble and expense where the only tax effect achieved would be to restore them to their position under the existing law.

45

Consideration was given to an amendment of the general deductibility rules which stopped short of interfering with the existing basis for dividends passing from subsidiary to parent or from one associated company to another. That gave rise to difficult issues about the degree of association which would qualify, and to complications which would be required to the legislation to ensure that other companies could not take advantage of the outlet. Furthermore it was judged that if the existing basis was preserved for associated companies and if life offices were put aside as a special case for later consideration, revenue collections from the remaining ordinary companies were not then much affected by the system.

In the upshot the legislation amended the law to deal with only two specific situations dividend stripping and the situation which arose from the 1965 decision in the Palvestments case. In that case a share trading company had expenses which, in the normal course of events, would have exceeded its non-dividend income, so that the excess would have been offset against dividends and would have reduced its section 46 rebate. The shares which it held were trading stock for the purposes of the income tax law, and the company revalued them to an extent that its non-dividend income was increased to a level at which it fully absorbed the company's expenses, and there was no loss of section 46 rebate. In effect, the company 'manufactured' income in that year by bringing it forward from a later year via the stock valuation provisions, thereby preserving an otherwise unused rebate entitlement in the earlier year.

In dealing with dividend stripping the legislation authorized the Commissioner, in forming a decision as to what deductions to attribute to dividends, to have regard to the transaction as a whole, including the cost price of the shares or any loss on their disposal. In circumstances where income had been 'manufactured' via the trading stock provisions so as to maximize artificially the section 46 rebate, the rebate was to be reduced to the amount that would have been allowed if the 'manufacture' of income had not taken place.

When introducing the amending Bill in 1972 to deal with dividend-stripping and Palvestment situations, the then Treasurer noted that the general section 46 amendment was not to proceed and said that 'it has become quite clear that there is no simple way of altering the basis of the rebate without giving rise to serious anomalies between different company structures'.

Subsequently, there was an amendment in 1973 to the calculation of the section 46 rebate on dividends received by life insurance companies, by the insertion of section 116AA in the Act. Prescribed portions of deductible expenses of general management and of the section 115 deduction for calculated liabilities were to be offset against gross dividend income in calculation of the rebate under section 46.

46

APPENDIX C3 - SECTION 46: ISSUES OF PRINCIPLE

In an attempt to clarify the correct conceptual treatment of deductions against dividends for rebate purposes, it may be helpful to begin with first principles and consider how dividends would be taxed under ideal arrangements.

The relevant principles are well known:

• the tax system should not impose different tax burdens on income which arises through different types of business organisation. In other words, the system should charge the same tax on an investor whether he undertakes his activity directly as an individual, via an incorporated entity, or via a series of incorporated entities;

• tax liability should be based on net income; ie gross income minus expenditure incurred in earning that income; and

• incremental net income should bear a total tax rate equal to the marginal rate appropriate to the individual beneficiary.

As noted in the draft White Paper, the ideal approach to give effect to these principles is full integration of the personal and company tax systems:

• there would be no company tax as such on company income but, as a withholding arrangement, companies would make a tax payment which was deemed to be a prepayment of personal tax on the share of total company income in which each shareholder had an interest;

• that payment of tax by the company would be at a flat rate, say equal to the current 46 per cent company rate; and

• individual shareholders would include in their personal tax returns not only their dividend, as at present, but also their share of the undistributed profits of the company. They would be taxed on both of those amounts, but would be allowed a credit for their share of the tax prepaid by the company. Those with marginal rates below the company rate would receive a net credit which could be offset against tax on their other income which could, in some circumstances, lead to a refund of part of the tax prepaid by the company.

The full imputation system proposed for introduction in 1987-88 will meet the broad intention of integration as applied to dividends, but not as applies to the undistributed income of companies. The appropriate tax treatment of dividends is illustrated initially in the context of such a system.

47

Full Imputation System

All examples below assume:

• that the hypothetical individuals and companies have sufficient other income to fully utilize any available interest deductions or excess imputation credits;

• full distribution of post-tax company income; and

• expenses comprise solely interest payments.

The exact manner in which the imputation system will operate as between various layers of companies is not important to the final outcome though some of the arithmetic varies. The examples below are presented initially with a system whereby imputation rolls through the various tiers of interposed companies. Under such a system, a section 46 rebate would not be required. Subsequently, consideration will be given to how the section 46 rebate would need to operate to provide the same outcome.

Table 1 takes as its starting point (case I) an individual investing directly in an income producing asset, yielding gross income of $500 per annum. It is assumed that the income producing asset cost $2000 which the individual purchased with $1000 of his own money and $1000 borrowed at 15 per cent. The average tax rate on net income is 30 per cent.

48

TABLE 1: FULL IMPUTATION SYSTEM

Gross Investment Income

$500.00

   

Interest Bill

$150.00

   

Company Tax Rate

49%

   

Personal Tax Rate

30%

   
       
 

CASE I

CASE II

CASE III

       

COMPANY A

     

Investment Income

 

500.00

500.00

Company Tax

 

-245.00

-245.00

       

Dividend Distribution

 

255.00

255.00

       

COMPANY B

     

Grossed-up Dividends

   

500.00

Interest Bill

   

-150.00

       

Taxable Income

   

350.00

Company Tax

   

-171.50

Imputation Credit

   

245.00

Net Credit

   

73.50

       

Dividend Distribution

   

178.50

       

INDIVIDUAL

     

Grossed-up Dividends

 

500.00

350.00

Investment Income

500.00

   

Interest Bill

-150.00

-150.00

 
       

Taxable Income

350.00

350.00

350.00

Personal Tax

-105.00

-105.00

-105.00

Imputation Credit

 

245.00

171.50

Net Credit

 

140.00

66.50

       

After-tax Income

245.00

245.00

245.00

       

Total Tax Paid

105.00

105.00

105.00

Legend:

Case I: Individual invests directly.

Case II: Company A earns investment income, and individual incurs interest and receives dividends from Company A.

Case III: Company A earns investment income and distributes dividends to Company B which incurs interest and has sufficient non-dividend income against which to off-set this interest. Company B in turn distributes dividends to individual.

49

Case II is the same as case I except that it is assumed that the individual invests through a company (company A) which uses the $2000 invested to aquire the income-producing asset. The ratio of total taxes paid to net income remains at 30 per cent; the absolute amount of total tax paid and the individual's after tax income are unchanged from case 1.

In case III, rather than investing directly in company A, the individual invests his $1000 in company B, which adds $1000 of borrowed funds to acquire shares in company A which purchases the income producing asset. Then company B would receive a gross dividend of $255, which would be grossed up to $500, pay interest of $150, and have taxable income of $350. This would be subject to company tax of $171.50; however, company B would be eligible for an imputation credit of $245, implying a net credit of $73.50.

Company B could then distribute $178.50 to individual shareholder, with comparable gross up and crediting procedures. In this situation, the individual's tax and after tax income would be identical to that in cases I and II.

This is an appropriate result, notwithstanding that in cases I and II the individual invested $2000 whereas in case III he apparently invested only $1000.

This is clear in the case where there is a close association between the individual and company B; say, where it is the shareholder's own private company. In such a situation, it is a simple matter to have the company do the borrowing instead of the shareholder himself. In these circumstances, cases 11 and III are substantively identical and the result above is clearly appropriate.

It remains appropriate in the unassociated (or many) shareholder case. The company is merely a vehicle through which the shareholders decide to conduct their business. The company is a conduit for transmitting the net earnings of the company into the hands of its shareholders. It is not only the gross income that belongs to shareholders but also the liability for the debt. The net income that should be taxed in the hands of shareholders is identical in all three cases above at $350.

Consider now the alternative situation where under imputation, the section 46 rebate is retained. Under this assumption, the imputation credit would not be 'rolling' as above, but simply be available at the time the cash dividend is received by the individual shareholder. Case III from Table I above is reproduced below (Table 2), with this alternative mechanism. It is clear that in order to achieve the same (appropriate) result that obtained under imputation, it is necessary for the rebate to be provided on the gross rather than the net dividend.

TABLE 2: FULL IMPUTATION SYSTEM

Gross Investment Income

$500.00

Interest Bill

$150.00

Company Tax Rate

49%

Personal Tax Rate

30%

   

CASE III

   

COMPANY A

 

Investment Income

500.00

Company Tax

-245.00

   

Dividend Distribution

255.00

   

COMPANY B

 

Interest Bill

-150.00

Tax Saving

73.50

   

Dividend Distribution

178.50

   

INDIVIDUAL

 

Grossed-up Dividends

350.00

Investment Income

 

Interest Bill

 
   

Taxable Income

350.00

Personal Tax

-105.00

Imputation Credit

171.50

Net Credit

66.50

   

After-tax Income

245.00

Personal Tax

 

Total Tax Paid

105.00

Legend:

Case III: Company A earns investment and distributes dividends to Company B which incurs interest and has sufficient non-dividend income against which to off-set this interest. Company B in turn distributes dividends to individual.

Therefore, under this form of imputation system, allowing the section 46 rebate on the gross dividend would (in cases where there are adequate amounts of non-dividend income) give the same outcome as the benchmarks of direct individual investment or direct investment in company A - benchmarks which provide the conceptually correct treatment of dividend income.

Classical System

Under the classical system, the situation is far less clear. The classical system lacks the coherent logic of the imputation system. Table 3 sets out three analogous cases to those shown in Table 1, plus an additional case IV (which has, however, lacked much empirical relevance for at least the past thirty years).

TABLE 3: CLASSICAL SYSTEM

Gross Investment Income

$500.00

     

Interest Bill

$150.00

     

Company Tax Rate

49%

     

Personal Tax Rate

30%

     
         
 

CASE I

CASE II

CASE III

CASE IV

COMPANY A

       

Investment Income

 

500.00

500.00

500.00

Company Tax

 

-245.00

-245.00

-245.00

         

Dividend Distributions

 

255.00

255.00

255.00

         

COMPANY B

       

Interest Bill

   

-150.00

-150.00

Tax Saving

   

73.50

 
         

Dividend Distributions

   

178.50

105.00

         

INDIVIDUAL

       

Investment Income

500.00

     

Interest Bill

-150.00

-150.00

   
         

Taxable Income

350.00

105.00

178.50

105.00

Personal Tax

-105.00

-31.50

-53.55

-31.50

         

After-tax Income

245.00

73.50

124.95

73.50

         

Total Tax Paid

105.00

276.50

225.05

276.50

Legend:

Case I:

Individual invests directly.

Case II:

Company A earns investment income, and individual incurs interest and receives dividends from Company A.

Case III:

Company A earns investment income and distributes dividends to Company B which incurs interest and has sufficient non-dividend income against which to off-set this interest. Company B in turn distributes dividends to individual.

Case IV:

Company A earns investment income and distributes dividends to Company B which incurs interest but has no non-dividend income against which to off-set this interest. Company B in turn distributes dividends to individual.

On neutrality grounds, case I remains a relevant standard of comparison. The fact that the classical system has some major non-neutralities does not mean that neutrality ceases to be an important consideration in coming to judgments on design details within that overall system.

Case II illustrates the situation where the individual invests directly in company A.

Case III illustrates where he invests via an interposed company (company B) and where the rebate is based on the gross dividend. Company B's interest bill then generates a tax saving in the same manner as in the corresponding case under imputation.

Case IV is as for case III but with the rebate calculated on a net basis. The $105 would be freed from tax and could be paid as a dividend to the individual. His taxable income would be $105, as in case II.

Case III is a more neutral outcome than 11 or IV and is more appropriate for investment through companies under our classical system. It produces a total tax take somewhat closer to that of the individual investing directly. More importantly, regardless of the level of debt used by company B to acquire indirectly the income producing asset, it limits the taxation of net income to double taxation, once at the company tax rate and once at the personal tax rate (resulting in a combined 64.3 per cent rate of tax on net income). Cases 11 and IV involve substantially more than double taxation.

Shareholdings Between Associated Companies

Furthermore, under a classical system, in less simple situations some further questions arise, including those relating to associated companies, which posed difficulties when the matter was under examination in 1971.

Consider first the simplest case of a parent company - company B - with a 100 per cent wholly-owned subsidiary company A - with transactions as per the next table (Table 4).

TABLE 4: ILLUSTRATIVE TAX LIABILITY OF ASSOCIATED COMPANIES: FULL AND RESTRICTED INTEREST DEDUCTIBILITY

   

Company A

Company B

(a)

Merger

     

Case I

Case II

 
   

$

$

$

$

1

Gross operating income

300

200(b)

200(b)

500

 

Dividend received

--

153

153

--

           

2

Interest deducted

       
 

against: operating income

--

160(c)

7

160

 

dividend received

--

--

153

--

           

3

Taxable income

300

193

193

340

 

Rebatable income

 

153

--

--

           

4

Tax(d)

147

94.6

94.6

166.6

 

less rebate

 

75.0

--

--

 

= net tax

 

19.6

94.6

166.6

           

5

Total post-tax income

153(e)

173.4

98.4

173.4

(a) Assumes 100 per cent common ownership.

(b) Own income (ie. excluding any transfers from company A).

(c) Equals actual interest payments.

(d) 49 per cent rate.

e) Assume paid fully to company B.

Case I is where the parent company B is allowed to deduct interest expenses in full against non-dividend income; case II is where the parent company is allowed to deduct interest expenses against non-dividend income only to the extent they exceed dividends received from the subsidiary.

Tax principles require that, on neutrality grounds (organisational structure should not affect the aggregate corporations' tax liability), their overall tax treatment be as achieved by case I treatment which effectively ignores the intra-group transfer. It is not achieved by case II treatment. With case II treatment, the tax liability is higher by $75. This is precisely equivalent to subjecting the dividend received by the parent to tax in their hands - ie to subjecting dividends to double taxation. Viewing company A and company B as a corporate whole, the intra-group transfer should be ignored for tax purposes. To do otherwise would be to introduce a bias into the system in favour of mergers (or other activities such as transfer of assets).

Appendix C2 describes some other problems raised in 1971 about a change to the basis of the section 46 rebate for dividends passing between associated companies. The example given in paragraph 11 of that Appendix is one that could, perhaps, have been dealt with fairly simply if it had been the only problem. It was a case where the parent undertook some services for the subsidiaries and charged them a fee to cover the costs.

However, more complicated situations would be the rule, where the parent did other things and allocation by anything other than an arbitrary formula would be more difficult. Where the parent company borrows on behalf of the group, which would be common, the interest expenses would be large. It would, no doubt, be possible for the subsidiaries to do their own borrowing, perhaps with the help of guarantees by the parent company. Under present law the rebates on dividends received by the parent company could be much the same, whoever did the borrowing. But if some of the interest expense were allocated to dividends there would be a significant loss of rebate if the parent did the borrowing.

As in 1971, there is reason to question whether that outcome should be brought about, as it would mean that the group would, purely for tax reasons, need to structure itself differently and still end up in the same after-tax position as if the change to the rebate provisions had not been made. If anything, the introduction of group loss transfer provisions since 1971 adds force to the view that group arrangements affecting section 46 rebates should not be disturbed.

ATTACHMENT D - OVERSEAS TAX EXPERIENCE

Canada

In Canada, prior to 1972 the interest expense incurred by a corporation for money borrowed to finance a share purchase was non-deductible. This provision was repealed as part of the tax reform package in 1972 for three main reasons:

• Since American corporations could deduct their interest expenses incurred in a takeover bid, the Canadian prohibition placed Canadian corporations at a disadvantage in bidding for control of Canadian corporations. This argument was particularly persuasive at the time because Canadians were becoming increasingly concerned about the high degree of foreign ownership in the Canadian economy.

• It was argued that the measure discriminated against hostile takeover bids since if the bid was friendly the acquiring corporation always had the option of acquiring the assets of the acquired corporation and the rule did not apply. There was a further more general point to this argument. If the acquiring corporation could defeat the section by purchasing assets (and no one argued that interest should be non-deductible in an asset purchase) in principle did it really matter that the interest expense was in another corporation if shares were purchased? Indeed, in many cases the section could be defeated by winding the acquired corporation up into the acquiring corporation once the takeover bid was successful and deducting the interest expense directly from the income-producing assets of the acquired corporation. Again, since this was always possible, should it matter what corporate income (or income in what corporation) is sheltered by the interest expense?

• A final argument that was made was that it was extremely difficult in many takeover bids to trace the use of the borrowed money. Obviously if the takeover bid was large this would not be so much a problem though, even then, legal questions could arise. However, in a substantial number of takeovers the acquiring corporation has enough flexibility to stream borrowed money into income producing uses and use assets to finance the takeover bid. Indeed, the effect of the tracing rule was to give an advantage to large corporations over smaller corporations in takeover bids.

United States

A provision (section 279) was introduced in the United States in the late 1960's which disallowed a deduction for interest in excess of $5 million per year on 'corporate acquisition indebtedness'. It was directed at a form of financing whereby promoters would make a public offering of convertible debentures and then use the resulting revenue to finance the acquisition of the selling corporation.

Under this provision, 'corporate acquisition indebtedness' is defined as debt to acquire stock or two-thirds of the non-cash operating assets of another corporation if:

• the debt is substantially subordinated (ie ranks behind other liabilities in the event of winding up and has some features in common with equity capital); and

• the debt is convertible into the stock of the issuer or carries an equity participation, such as warrants to purchase stock of the issuer; and

• the issuer is thinly capitalized (debt-equity ratio exceeds 2:1 or the projected earnings do not exceed three times the interest on the acquisition debt).

The provision is easily avoided, and the IRS rarely invokes it, because it applies only if the indebtedness satisfies all of the requirements above. It is seldom difficult to avoid at least one of the requirements. The requirements themselves are complex (there are associated complicated regulations under section 279 spelling out the meaning of many of the terms) precisely because of the desire to limit the scope of the provision ie to ensure that it only applies where the debt used to finance the proposed takeover is a very close substitute for equity. As it is a relatively simple matter for the acquiring company to use other debt obligations not covered by the provision, the latter is of little practical relevance in takeover situations.

The US Congress has examined the relationship between taxation and mergers from time to time. Most recently, during 1985, both the House Ways and Means Committee and the Senate Finance Committee held hearings on the tax treatment of hostile corporate takeovers. No tax related changes to the law have emerged from these examinations.

More recently, academic debate in the US has centred on the argument that it was impossible for mergers to be tax driven because the tax advantages available to the acquiring company are also available to the target company. Possibly partly for this reason and because debate on relevant tax angles has been swamped by the on-going consideration of the general US tax reform package, the chief focus of the current debate in the US on takeovers is on financial prudence and management aspects of takeovers.

There are also general provisions in the US revenue code which lay down guidelines (including in relation to debt-equity ratios) for distinguishing debt from equity in order to determine whether interest on corporate borrowings should be disallowed where the relevant debt instrument is considered to be more akin to equity. The aim of these guidelines is generally to ensure that the revenue does not bear the cost of deductions for interest payments that are effectively dividends on a disguised equity issue; the rules are not confined to takeovers but are applicable in all situations arising under the tax law where there is a need to distinguish debt from equity for income tax purposes. The US guidelines/rules in this area are complex and attempts by the revenue authorities to issue regulations designed to clarify the relevant law have not yet been successful - in the absence of such regulations the position is currently governed by case law. Moreover, we understand that the grounds most frequently relied on for attacking purported debt as equity do not apply to most highly geared takeover transactions because the debt instruments used to finance the offer generally fall outside the scope of the guidelines.

It might also be noted that section 246A (which was introduced by the 1984 Tax Reform Act) of the US code provides that in certain circumstances the 85 per cent dividends received deduction on intercorporate dividends is reduced for dividends received from debt financed portfolio stock - the provision generally reduces the deduction for dividends received on such stock so that the deduction is effectively available only with respect to dividends relating to that portion of the stock which is not debt financed. The measure only applies in relation to defined portfolio investments in shares, and not where the company making the investment has a 50 per cent or greater interest in the other company.

The measure was apparently designed to discourage certain tax minimization practices based on the issue of preference stock by a tax loss company to a taxable company which could obtain the benefit of the dividends received deduction on the preference dividends. In such a situation there were tax advantages available to both parties through the issue of preference stock rather than a debt instrument. There were also other tax advantages available to the purchaser where the latter sold the preferred stock after receipt of a cumulative dividend payment and offset the resultant loss against capital gains derived on other investments.

There is some doubt as to the effectiveness of the measure because the relevant portfolio indebtedness has to be directly attributable to investment in the shares and the provision does not incorporate any allocation or apportionment formula or fungibility concept. The provision only applies if indebtedness is clearly incurred for the purpose of acquiring dividend-paying portfolio stock or otherwise is directly traceable to such an acquisition. It seems that the provision is destined to encounter all of the practical problems that we have faced in relation to the operation of the section 46 rebate provision and the initial response of US practitioners is consistent with this.

Furthermore, the measure was not specifically designed to discourage mergers or takeovers although it could act as a disincentive in the early stages of a proposed bid. It seems that the section is directed more at the possible tax advantages available after a takeover has been completed than the takeover itself.

United Kingdom

The debate over takeovers in the United Kingdom appears to be a good deal behind that in Australia and the United States. Apart from the usual thin capitalization rules applicable to foreign investment funded from foreign non-arm's length sources, there do not appear to be tax-based provisions in UK law relating to highly geared takeovers.

Interest paid by an individual is generally not deductible in the UK. Exceptions relate to certain investments in private companies. Thus, interest paid on borrowings to acquire shares in a public company are not deductible, even though the related dividend income is assessable in the hands of the individual.

On the other hand, interest paid by UK companies, including the acquiring company in a public company takeover, is generally deductible for UK tax purposes. Any excess deductions cannot normally be carried forward to future years but it can be transferred to other companies in the same group. More generous treatment applies, however, to 'investment companies'.