SUPPLEMENTARY REPORT : SENATOR ANDREW MURRAY AUSTRALIAN DEMOCRATS : November 1999

Introduction

This inquiry has been specifically about the impact of business tax reform on the Federal Budget. The Government has made an assertion in its media statements [1] that business tax reform will, despite being revenue negative in 2001-02, be revenue positive over its first five years (2000-2005), to the tune of around $1.27 billion.

For any forecaster, the estimation of revenue up to five years into the future is fraught with difficulty and uncertainty. Indeed Treasury witnesses to this committee have pointed out that revenue estimates, even a year into the future, have a considerable margin for error. Government revenue figures are of such a magnitude that such errors in forecasting can translate into billions of dollars, and the Treasury relies on 'overs and unders' to balance the effects.

Each costing exercise of revenue gained or revenue foregone involves the making of assumptions about its likely effect. This inquiry has been unable to test each and every assumption. But, it has been able to take evidence on a number of key issues. In this Supplementary Report, I seek to briefly outline the Democrats' views on some of these key issues.

The Democrats do not believe that the business tax package, as it stands, will be revenue neutral. We have concluded that the behavioural assumptions on the capital gains tax cuts provide for an over-optimistic estimate of likely revenue gains which knocks more than $2.5 billion out of the forward estimates, leaving the Budget in the red by around $1.49 billion over five years.

We also conclude that the capital gains tax cuts in their current form are likely to lead to considerable tax avoidance, more than that allowed for by Treasury.

We are also concerned that some other measures may provide a risk to revenue greater than that admitted to by the Ralph Report, and the Government.

And, we conclude that in the longer term, the 'growth dividend' may not plug the very considerable hole in the forward estimates beyond five years, as the revenue effects of changes to accelerated depreciation wash out.

Against that, the Democrats recognise that Treasury has been conservative in its estimates of revenue gains from some measures, such as the 'growth dividend', the personal services income initiative, and the 'Option 2' higher level reform. On two of these matters, because the final legislation has not been drafted, and further concessions and exemptions may be granted between the announcement and the passage of the final law, it is difficult to comment on the fiscal effects. On the 'growth dividend' issue, it is prudent to be conservative.

The importance of revenue neutrality

The 2000-01 and 2001-02 Budgets are likely to be very tight. This is because of three major factors which kick in early in the tax reform programme:

the cost of the personal income tax cuts and the ANTS package;
the cost of the compensation measures for the introduction of the GST; and,
the cost of the commitment to East Timor in terms of defence, UN administration and aid programs (up to $2 billion according to press reports).

The OECD has warned that maintaining the Budget in surplus is essential, given Australia's large current account deficit, to maintain financial market confidence. [2] Similarly, the Reserve Bank has warned that the prospective reduction in the Commonwealth's fiscal surplus in 2000-01 is a factor that supports growth, at a time that the Bank is moving to raise interest rates to `move monetary policy to a less accommodating position'. [3] Australia should not allow the Budget to be pushed into deficit with the risk of pushing interest rates higher than they might otherwise be.

Access Economics, in its most recent Budget Monitor, has warned that tax cuts and spending will push the Budget balance from a $5.2 billion cash surplus this year into a $500 million cash deficit next year. They warn that:

The bottom line is that the economy is delivering less to the Budget revenues than many may have expected, and that Government spending is cutting larger chunks from the surplus than many realise. [4]

Other bank economists disagree with Access' estimates, and forecast a surplus of up to $2 billion. These varying views should encourage the Government to err on the side of caution. As Access points out:

That is a reminder of the importance of revenue neutrality for the Ralph measures - like the GST they are necessary reforms, but they should not come at too high a cost in taxpayers' money. [5]

An expansionary fiscal policy providing business tax cuts above those of the personal income tax cuts is simply not responsible. Business tax must pay its way and, if possible, contribute additional revenue to the Budget. For these reasons, the Democrats believe that conservative assumptions should be used in calculating business tax revenue effects to minimise the prospect of a Budget deficit blowout, and to maximise the prospect of a Budget revenue windfall in what would others be a very difficult fiscal year.

Capital gains tax revenue - impact of elasticities

As outlined in the main committee report, the key weakness in the revenue figuring behind the Ralph Report lies in the Capital Gains Tax proposals, particularly in the treatment of realisations.

The Democrats conclude that Ralph was over-optimistic in assuming that a cut in capital gains tax would deliver an elasticity of 1.7 in the short term and 0.9 in the long term. The more recent research projects referred to by Dr Gravelle and Professor Auerbach in the evidence show quite clearly that these estimates are far too generous. Dr Gravelle, who outlined in detail her conclusion that the short-term elasticity is likely to be closer to 0.6 and the longer term to 0.2, concluded that:

it means that this tax cut, which I understand has been presented as a revenue neutral tax cut, is in fact going to lose revenue, if that elasticity is correct. It will be quite a big difference. [6]

Private sector economist Dr John Edwards, in his submission to the inquiry, said that:

I frankly doubt we would see it on anything like the scale implied by the elasticity estimate. [7]

The Democrats believe that, in erring on the side of caution, the elasticity estimates used should be those proposed by Dr Gravelle rather than those adopted in the Ralph Report. This affects the capital gains tax revenues in three ways:

it reduces the revenue gain flowing from the personal income tax cut;
it reduces the revenue gain flowing from the super funds tax cut; and
it reduces the revenue gain flowing from the scrip-for-scrip rollover relief.

The revenue impact of these three effects are summarised in the following table:

Impact of lower elasticity figure on capital gains tax realisation estimates

$m	00-01	01-02	02-03	03-04	04-05
Personal CGT cut - Ralph claim of realisation	540	530	500	480	400
Super funds CGT cut - Ralph claim of realisation	70	50	40	40	30
Scrip-for-scrip relief - Ralph claim for revenue	30	0	10	20	40
Personal CGT cut - lower elasticity*	198	180	58	62	60
Super funds CGT cut - lower elasticity#	25	18	5	5	5
Scrip for scrip relief - revenue claim with lower elasticity@	n.a.	-48	-55	-60	-58
Total reduction in revenue estimates	487	480	542	533	463

* This is used by calculating the static tax cut by 0.6 for the first two years and 0.2 for the subsequent years.
# Because a lower elasticity is used by Ralph for super funds, the revenue estimate is the same proportion of the Ralph claim as for the personal income tax cut.
@ This is worked out using the sensitivity tables in the Access Economics report on scrip-for-scrip relied on by the Ralph report for its 70% (or 0.7) realisation figure. In the first year (2001/02), the elasticity of .6 is used instead of 1.25, and in later years .2 is used instead of .7.

The generosity of the lower capital gains tax is also likely to lead to an increase in tax avoidance, particularly because of the retention of a 100 per cent deduction for negative gearing despite only a 50 per cent capital gains tax. Professor Rick Krever, economist Dr John Edwards and US proponent of capital gains tax cuts, Alan Reynolds, in evidence, warned that this combination was likely to lead to an increase in tax avoidance. Dr Edwards warned that negative gearing was in fact on the rise in the equities market:

negative gearing is now, even with the current capital gains tax, a hugely popular strategy. In most banking institutions, for example, marginal lending into equities is the most rapidly growing part of their business. You can see it in the housing finance figures. There are huge increases in individual credits lent on the security of houses which are going into the stock market or into other properties. [8]

The Ralph Report, in dealing with the ability to negatively gear non-commercial losses, originally carved out a continuing exemption for rental properties, but not shares. [9] The Government, in its Stage 2 response, broadened this out to also include shares, without any indication of why, or any adjustment of the revenue costing.

Professor Chris Evans in evidence suggested that even a slight movement in tax planning from earned income to capital gains would reduce the revenue take considerably. He found that a one per cent movement of converting income to capital would cost $359 million in annual revenue, based on 1996-97 figures. [10] In 1996-97, negative gearing cost the tax system around $937 million in tax foregone on losses on rental properties alone. [11] Just a ten per cent increase in negative gearing would cost around $94 million a year. Professor Krever and Dr Edwards also warned that the tax arbitrage effect was likely to be higher than predicted by Treasury.

Given these sorts of risks, the Democrats believe that the tax arbitrage effect allowed for in the Ralph estimates of a maximum of $180 million a year in lost tax due to conversion of income to capital is likely be an understatement given the size of this capital gains tax cut. For the purposes of this exercise, we believe it would be prudent to increase the revenue lost allowance by at least 50 per cent.

The result of the changes to the realisation and tax arbitrage effects are included in the table below. The effect is to convert the capital gains tax measures from being net revenue positive to net negative, and the package as a whole from net positive to net negative.

Overall effect of correcting capital gains tax estimates

$m	00-01	01-02	02-03	03-04	04-05
Capital Gains Tax measures – Govt claim	132	151	85	41	-46
Capital gains tax measures – Democrat estimate	-365	-354	-507	-567	-599
Total package fiscal effect – Govt claim	707	-525	411	559	115
Total package fiscal effect – Democrat estimate	210	-1030	-181	-49	-438

The Democrats conclude that the capital gains tax gains to revenue have been overstated and that the loss to revenue has been understated. By correcting for these effects, we conclude that the package, rather than being revenue positive by $1.267 million as claimed by the Government, is in fact revenue negative to the tune of $1.49 billion.

The Democrats will have to determine how to attempt to ameliorate problems arising from both the real cost of the proposed individual capital gains tax cuts, and some equity considerations outlined in evidence to the committee.

Other revenue risks

Wind-down of accelerated depreciation

A number of witnesses, including Dr Matt Benge and Geoffrey Lehmann, warned that the savings from abolishing accelerated depreciation are not permanent, and will phase out over time as the offsetting cost of allowable delayed deductions are paid out. The Ralph Report figures have the savings peaking at $2.61 billion in 2003-04 and falling off slightly the following year to $2.55 billion, and probably further in subsequent years. Mr Lehmann, while acknowledging the problem, predicts that increased growth rates and the conservativeness of the estimates on the revenue gains from tax integrity measures should `easily accommodate a reversal of the extra revenue from eliminating accelerated depreciation'. [12]

This may be so, but it highlights the importance of the difficult-to-estimate 'growth dividend', which emphasises the need for the package as a whole to be revenue neutral or, indeed revenue positive, in its early years.

Growth dividend

The Ralph Report (pages 728, 698 and 19-26) forecasts a very high 'growth dividend' flowing through from the tax changes. This growth dividend is essential to the package being revenue neutral over the medium and longer term. The Report makes a judgement that the reforms would deliver a growth dividend of 0.75 per cent of GDP by 2009, which the report says is `likely to be conservative'. [13] This would deliver a revenue gain by 2004-05 of $650 million, which is rounded down to a `conservative' $500 million in the Ralph Report tables. The Government, in its official press releases, halves this figure again to a $250 million 'growth dividend' in 2004-05. However, as highlighted earlier, the importance of the growth dividend is in the medium term to offset the winddown of accelerated depreciation. The Ralph Report claims that its 0.75 per cent dividend would push the revenue gain out to $1.8 billion by 2009-10.

Chris Murphy's modelling supported the view that there would be a growth dividend in the long run (i.e. 10 years), which would be a boost of 1.5 per cent to GDP, and concludes that there would be a `substantial growth dividend out of all that in terms of Government revenue'. [14]

However, Dr Matt Benge [15] and Professor Peter Dixon [16] disagreed with this assessment, coming to the tentative conclusion (on very limited data), that the overall effect of the package was likely to be negative on new investment and hence on long term GDP growth.

Long run modelling is very difficult to make firm conclusions on, particularly given the limited material in front of this inquiry. The Democrats would be reluctant to rely on high assumptions of revenue gains from a 'growth dividend' which may or may not arrive. We note that the Government is running on a growth dividend less than 40 per cent of that forecast by Ralph. We commend their conservatism in this regard. The reliance on the growth dividend to fund the package in the medium to long term highlights the importance of ensuring that the package is robustly revenue neutral or indeed revenue positive in the short to medium term.

Treatment of loss-carrying entities in group consolidation

The treatment of loss carrying entities in the process of consolidating groups is the largest new revenue concession in this package aside from the company tax cut, capital gains tax cut and the small business measures. This losses measure peaks in cost at $390 million in 2003-4, and costs $1.26 billion in the four years 2001-05. No evidence was entered on this item, but it does need to be noted as a significant potential revenue risk.

The consolidation of company groups should, in theory, result in an increase in revenue. By treating all related companies as a single entity, artificial losses between them are discounted. But, amazingly, the group consolidation process is going to cost revenue - around $200-$290 million a year because the consolidation process will allow groups to acquire loss-carrying entities and to claim their losses against income earned by other entities in the group. This will allow a faster release of the huge bank of corporate tax losses ($44.6 billion of revenue losses and $21 billion of capital losses) to be offset against the taxable income of other parts of the groups. The February Ralph discussion paper proposed six options of progressively increasing conservatism for dealing with losses in groups: [17]

not allow losses into the group, unless they fit the current tight transfer rules which cover 100% owned;
losses could be transferred if the entity and Group are continuously owned since the loss was incurred (or satisfy the same business test), but with losses transferred over a 10 year period;
as in 2, but with the further rule that the loss can only be transferred if the entity would have been able to use the loss within a specified period (eg. 1 year) if it had not been consolidated;
allow losses to be transferred according to the proportion that the group owned the entity, with any remaining loss proportion lost;
quarantine losses from the entity against income from the same entity within the group for say seven years; and,
leave loss making entities outside the group until the losses are recovered.

The July Ralph Report recommended a hybrid of these:

if the entity and the group satisfy the `continuity of ownership' test, then the proportion of losses relating to the group's ownership when the loss is incurred be brought in immediately, and the remainder over five years;
if it does not satisfy the ownership test, but satisfies the `same business test' (SBT), provided the losses are less then $10 million or 5 per cent of the equity in the entity, the losses be brought in over 5 years; and
in any other case, the equity can be brought in with the SBT cap applied to it, or leave the entity outside the group until the SBT cap is satisfied.

This treatment is clearly more generous than that proposed in February paper. The Democrats are concerned at the generosity of the relief proposed, and question whether, given the difficult Budgetary situation, that revenue risk should be taken. We question whether one of the less generous options (e.g. not consolidating an entity until its losses are expunged) should be adopted, and whether the item could end up being considerably more expensive than claimed given the huge store of corporate losses. Clearly, this proposal carries serious risks and a cautious approach should be adopted.

Timing of tax avoidance measures

The Democrats welcome the decision of the Government to proceed with the Stage Two Tax Integrity Measures with full effect from July 1 2000. We also note the determination of the Government to proceed with the entities taxation scheme (i.e. taxing trusts as companies) from July 1 2001. These items are very important to the funding of this package and the ANTS package. Any watering down or further deferral of their commencing dates will have serious implications for the revenue neutrality of this package. The amount of revenue involved is quite considerable - the Tax Integrity measures in Stage I and 2 total $1 billion in 2000-01, rising to $1.49 billion by 2004-05. Similarly, the trusts measure was estimated to raise $900 million in its first year, falling to $430 million by the third year (although these figures assumed a 36 per cent company tax rate). If any of these measures are watered down considerably, the capacity of the Budget to fund a $3 billion company tax rate cut is brought into doubt.

This package is a package, and the Democrats believe that the legislation for all of the measures should be considered as a package. While we appreciate that the task of drafting the personal service income and trusts legislation is daunting, we believe that such legislation should be considered and passed before the company tax rate cut takes effect on July 1 2000. Indeed, the one should be contingent on the other.

Further tax avoidance measures

As indicated above, the Democrats believe that this package fails the test of revenue neutrality in the short term, and carries additional risks in the longer term. The Government faces two options in dealing with this - it can reduce the generosity of the concessions on offer, or it can seek other alternatives for revenue. This is not the appropriate place for a detailed discussion of all the alternatives. However, we do propose to briefly list a small number of alternatives.

Reducing the FBT concessional treatment of company cars

Treasury estimates the concessional treatment costs upwards of $700 million a year. The Ralph Report itself acknowledged this. In its February report it said:

These concessional arrangements provide a strong incentive for some employees to take a car as part of their remuneration package and so skew their consumption toward motor vehicle services.

Ralph made a recommendation to reduce the concession. This should be done, but, given the transitional problems the car industry is having with the phase-in of the GST and input tax credits, the measure should not take full effect until the GST tax cut and input tax credits are fully phased in 2002.

Modifying the 'mutuality' principle and taxing the 'super' clubs as businesses

The 'super' clubs compete directly against hotels and restaurants for trade, but have the added advantage of tax exemption from income tax. Removing this exemption from the large clubs that operate on a commercial basis would save upwards of $200 million a year. [18]

Reducing negative gearing

The cost of negative gearing to revenue is enormous - $937 million in 1996-7 on rental properties alone. This inquiry has heard that cuts to the capital gains tax rate are, when aligned with negative gearing, likely to lead to considerable further revenue erosion. Halving the negative gearing concession in line with the 50 per cent capital gains tax rate would save around $470 million a year. Few other countries in the world (including the US, Canada and the UK) still allow negative gearing. Negative gearing in our opinion, promises to be an even greater revenue loss in the future, taken in conjunction with the capital gains tax cuts.

Proceeding with the 'Option 2' reforms of closely aligning taxable and reportable profits

The Democrats believe that the Option 2 reforms have considerable scope to reduce corporate tax avoidance and tax planning. It is impossible to estimate this amount, but we note that the Alternative Minimum Company Tax in the United States (which operated on a similar principle) increased revenue collections by 7 per cent in its first year.

Additional costs

The cut in the company tax has the undesirable (and perhaps unintended) consequence of cutting the effective tax concession on research and development from 9 cents in the dollar to 7.5 cents in the dollar. Geoffrey Lehmann, a witness to the inquiry, has calculated that to restore the R&D concession to its real incentive rate, would require an increase in the concession from its present 125% to 130%, at an annual cost to revenue of $97 million. A restoration of this desirable tax concession on this basis is obviously only possible if the package can be made revenue neutral.

A second issue is the question of maintaining or assisting viability for major projects such as those outlined in the submission by the Australian Gas Light Company. This whole cost area remains very vague in Government announcements to date, but with the loss of accelerated depreciation, plainly needs more clarity for such businesses.

A third issue is the delay until 1 July 2001 in introducing the 'Simplified Tax System' for small business. Ideally those small businesses that wish to should be given the option of moving onto that beneficial new system from 1 July 2000. There may however be too great a cost to revenue in bringing forward this measure. However it seems foolish not to allow new small businesses that start up in the year 2000-01 to go straight onto that system, rather than start with one system and have to change to a better system less than twelve months later.

Andrew Murray
Senator for Western Australia

Footnotes

[1] Treasurer, The New Business Tax System: Stage 2 response, 11 November 1999.

[2] OECD Economic Outlook, quoted in the Australian Financial Review, 17 November 1999, p. 4.

[3] Reserve Bank, Semi-Annual Statement on Monetary Policy, November 1999, p. 1.

[4] Access Economics, Budget Monitor, No. 43, November 1999, p. (i).

[5] ibid., p. iv.

[6] Evidence, 11 November 1999, p. 84.

[7] Submissions and Documents, p. 180.

[8] Evidence, 12 November 1999, p. 221.

[9] Review of Business Taxation, A Tax System Redesigned, 1999, p. 295.

[10] Submissions and Documents, p. 3.

[11] Taxation Statistics 1996-97.

[12] Submissions and Documents, p. 95.

[13] Review of Business Taxation, A Tax System Redesigned, 1999, p. 21.

[14] Evidence, 12 November 1999, p. 202.

[15] Evidence, 11 November 1999, p. 123.

[16] Evidence, 12 November 1999.

[17] A Platform for Consultation, pp. 561-4.

[18] See Senate Hansard, 18 October 1999, Senator Andrew Murray Adjournment Speech.

SUPPLEMENTARY REPORT : SENATOR ANDREW MURRAY AUSTRALIAN DEMOCRATS : November 1999