Briefing for the Incoming Minister of Revenue - 2005 - Part 2
Key challenges in tax policy
New Zealand has relatively broad tax bases for both income and expenditure and relatively low statutory tax rates. This might be thought to imply that, at least for households without children or beyond the income levels at which family assistance is typically abated, the New Zealand tax system is likely to work well and efficiently.
Nevertheless, there appear to be emerging and conflicting pressures. First, globalisation is placing downward pressure on the company tax rate, and many countries have reduced their company tax rates in recent years. New Zealand needs to decide whether it too should reduce its company tax rate. Second, there seems to be growing evidence of tax sheltering and income splitting, raising questions about the robustness of New Zealand's tax system. This appears to be at least in part a consequence of New Zealand's company and trustee tax rates being lower than the top marginal tax rate. By itself, any reduction in the company tax rate would add to the possibility of further tax sheltering.
Globalisation and downward pressure on company and personal tax rates
In an open economy like New Zealand's the company tax rate has a dual function.
First, for domestic shareholders it provides a withholding tax on income derived through companies. At present, the company tax rate is 33%, and the top marginal personal tax rate is 39%. So long as the company tax rate is less than the top personal tax rate, high-income individuals who wish to accumulate savings for a period before spending them on consumption will be better off if they hold their investments in a company than if they hold their investments directly. So long as profits are accumulating they are taxed at the company tax rate rather than the higher tax rates of shareholders. This can make companies a vehicle for sheltering income from higher rates of personal tax, even though when profits are eventually distributed they will be subject to tax at the tax rates of shareholders.
Second, for foreign shareholders and for domestic tax-exempt entities, company tax acts as a final tax. As a result of New Zealand's foreign investor tax credit (FITC) system, there is normally no additional tax when profits are distributed to foreign shareholders.
If not for international considerations, there would be a strong attraction to aligning the company tax rate and top personal rate, as was the case when New Zealand's full imputation reform was initially introduced. This prevents companies from being used to shelter income. Income sheltering results in people in the same circumstances (one with a company and one without) paying significantly different levels of tax. Such sheltering undermines a progressive tax rate policy and creates unfairness, thereby eroding the overall integrity of the tax system.
Nevertheless, the company tax rate also needs to be set taking into account international considerations. In particular:
- Too high a rate may discourage efficient inbound equity investment.[15]
- Higher rates than those of trading partners may create incentives for tax planning by transfer pricing and thin capitalisation to stream profits to lower-tax-rate countries.[16]
- Too high a company tax rate may increase distortions to investment decisions and constrain economic growth.[17]
An important phenomenon over the last 20 years has been a worldwide downward trend in company tax rates. Figure 8 shows the trend in the statutory corporate tax rates since 1985, excluding state taxes and surtaxes.[18]
 |
| Click on image above to get a full size view |
Between 1985 and 2004, the OECD average federal company tax rate fell from 43.5% to 31% on a GDP-weighted basis (or from 41.8% to 26.6% on an unweighted basis). Over the same period the Australian company tax rate fell from 46% to 30%, and the New Zealand rate fell from 45% to 33%.
At the beginning of the 1990s New Zealand's corporate income tax rate was well below the OECD average and similar to the average of a group of ten ASEAN countries.[19] Since then, reductions in the company tax rate in other countries mean that New Zealand's company tax rate is no longer relatively low.
Our relatively high company tax rate, at present, may make New Zealand an unattractive place in which to establish a regional headquarters. The importance of this is difficult to measure, and other considerations - such as size of market, other taxes, labour costs and other regulatory considerations - may often be more important.
A relatively high company tax rate also exposes New Zealand to transfer pricing and thin capitalisation pressures. Tax authorities in New Zealand and other countries endeavour to constrain transfer pricing and thin capitalisation but there is strong international evidence that such measures are less than perfectly effective. A particular area of vulnerability for New Zealand is Australia's tax system, given the importance of Australian investment into New Zealand. (Forty-six percent of foreign direct investment into New Zealand in 2003-04 came from Australia.) In 2000 Australia had a higher company tax rate than New Zealand, 36% compared with our 33%. Since then Australia has lowered its tax rate to 30%.
An additional consideration is that Australian shareholders in Australian firms gain imputation credits for Australian taxes, but not for New Zealand taxes. Even if the New Zealand and Australian company tax rate were aligned, there would be an incentive for New Zealand subsidiaries of Australian parent companies to stream profits to Australia. Stemming these incentives may require New Zealand to have a significantly lower company tax rate than Australia's.
At present, company tax contributes more than 15 percent of total tax collections and therefore contributes significantly to the government's overall fiscal position. The possible erosion of this tax base is a material fiscal risk. It is of interest to note that none of the five countries with greater tax collections as a percentage of GDP than New Zealand have higher company tax rates. A key concern is whether we will be able to continue to collect as much company tax as we do at present if we do not lower the rate of company tax. On the other hand, in the absence of any changes in corporate behaviour, lowering the company tax rate clearly reduces revenue.
Robustness of personal income tax system
In any tax system with varying marginal tax rates and a company or trustee tax rate below the top personal marginal tax rate there will be incentives to split and shelter income. By international standards, New Zealand has relatively little variation in statutory marginal tax rates. The range of 19.5% to 39% (or 15% to 39%, including low income earners' rebate) is relatively narrow.
Nevertheless, a number of key methods are currently used to split or shelter income. There are a number of specific anti-avoidance rules targeted at income splitting or sheltering as well as a general anti-avoidance rule, although their effectiveness is limited.
Companies
The use of a company is often a part of an arrangement that results in lower effective tax rates for individuals.[20] The business is undertaken by the company. The company income is taxed at 33% and when distributed is taxed again at the recipient's marginal tax rate, with an imputation credit for the company tax paid. If the company is owned by a shareholder whose marginal tax rate is 39%, the effect is one of deferral until the income is distributed.
Trusts
The income of a trust can be taxed as trustee income at a rate of 33% and later distributed to beneficiaries on a 39% marginal tax rate, with no further tax impost. Therefore the 39% marginal tax rate can be avoided by the use of trusts. Alternatively, the income can be directed to a beneficiary (perhaps a spouse) whose income may be taxed at a rate below 33%. It seems reasonable to assume that little income passed through trusts is taxed at 39%.
The use of a trust owning a trading company has become much more common recently, in which case the 33% company tax is often effectively a final tax, even though the beneficiaries of the trust may be taxed at a rate of 39%. Other assets are also frequently placed in a trust.[21] If the assets are income-earning (for example, as business premises can be) a 33% maximum tax rate is often achieved. Less commonly, but prevalent in some parts of rural New Zealand, trading trusts are being used for active business and achieve the same effect.
Families and associated persons
The personal income of a high marginal rate taxpayer can be split with an associated low marginal rate taxpayer to avoid tax. This might be achieved by transferring an income-producing asset to the low rate taxpayer. Business income might also be split by paying a wage to the low rate taxpayer, or by operating as a partnership and paying the low earning 'partner' a proportion of the profits. If such payments are not commensurate with the contribution of the low rate taxpayer the arrangement may be subject to general anti-avoidance provisions, although they will never prevent all income splitting.
SSCWT and FBT "salary sacrifices"
We have some evidence that "salary sacrifices" for employer superannuation contributions are being actively marketed. (A "salary sacrifice" is the forgoing of salary and wage income to get other benefits.) A salary sacrifice for employer superannuation contributions, which are taxed as SSCWT (specified superannuation contribution withholding tax), can result in salary income that would be taxed at 39% being taxed at a lesser rate. SSCWT involves contributions being taxed at 33% or even less.
There are also incentives for salary sacrifice for fringe benefits because of the non-abatement of family assistance against fringe benefits. As charities are not taxed on fringe benefits, there are likely to be particularly strong incentives for charities to remunerate their employees by way of fringe benefits rather than salary and wages.
Empirical evidence
Figure 9 shows aggregate income of individuals for the years 1999, 2002 and 2004. It shows three large peaks. The first is around the $9,000 to $18,000 area and relates to recipients of government transfer payments. The second and third peaks are around the marginal income tax thresholds of $38,000 and $60,000.
From 1989-90 to 1999-00 the top personal marginal tax rate was aligned with the company tax rate and trustee tax rate. From 2000-01 the top personal marginal tax rate was raised to 39%, while the company tax rate and the trustee tax rate were left unchanged. Since then there has been substantial growth in the number of taxpayers earning income around the 33% and 39% thresholds of $38,000 and $60,000, which is likely, at least in part, to be evidence of tax sheltering or income splitting.
 |
| Click on image above to get a full size view |
Figure 10 shows the growth in numbers of individuals with incomes in different income bands between 1998-99 and 2003-04.[23] In the absence of tax considerations, it might reasonably have been expected that there would be negligible or negative growth in the lowest income bands and strongest growth rates in the highest income bands as income growth shifts individuals up the income spectrum. However, numbers of taxpayers with incomes below $38,000 grew by 1.0 percent, those with incomes between $38,000 and $60,000 grew by 48.4 percent, those with incomes between $60,000 and $100,000 grew by 71.8 percent, those with incomes between $100,000 and $150,000 grew by 32.1 per cent, those with incomes between $150,000 and $500,000 grew by 24.7 percent, those with incomes between $500,000 and $1 million grew by 8.2 percent and those with incomes above $1 million fell by 6.3 percent.
While there is no obvious evidence of people sheltering from the top 39% marginal tax rate in the $60,000 to $100,000 range, the slower or negative growth rates for higher income individuals is likely evidence that significant numbers of those with substantial amounts to gain through income splitting and tax sheltering are availing themselves of these opportunities.
Between 1998-99 and 2003-04 the proportion of shareholder/employees reporting salaries of exactly $60,000 (to the nearest dollar) rose from 0.45 percent to 4 percent of such individuals. The number of individuals deriving over $100,000 from shareholder salaries almost halved over the same period, to 3,600.
 |
| Click on image above to get a full size view |
In the absence of tax considerations, there would be no a priori reason for there to be different growth rates for tax paid by companies, trusts and other persons. In the absence of any changes in behaviour, it might perhaps be expected that the increase in the top personal marginal tax rate would increase tax collections from other persons relative to trusts and companies. However, between 1998-99 and 2003-04 tax paid by trusts grew by more than 105 percent, to $480 million, while company tax grew by 71 percent, to $6,581 million, and tax paid by individual other persons rose only 16 percent, to $3,858 million.
Similarly, and in the absence of tax considerations, there would be no reason to expect different growth rates in the fraction of the income of trusts that is taxed as trustee income and the fraction taxed as beneficiary income. However, the increase in the top personal marginal tax rate to 39% means that those on the highest marginal tax rate can shelter income in trusts by having it taxed at a rate of 33% as trustee income.
Figure 11 shows trust income between 1998-99 and 2003-04, split by trustee income and beneficiary income. Over the five years from 1998-99 to 2003-04[24] trustee income grew by 256 percent and beneficiaries' income by 28 percent. Practically all of this income growth is in the form of imputed dividends.
 |
| Click on image above to get a full size view |
This difference in growth rates may not all be an indication of people sheltering from the top personal marginal tax rate. Part of the reason for the difference may be changes to the minor beneficiary rule in the Income Tax Act, which resulted in income distributed by a trust to a minor beneficiary being taxed as trustee income in 2002 and subsequent years.
Although there is no hard evidence yet of significant increases in salary sacrifice, there is strong anecdotal evidence that schemes using SSCWT to lower tax payments aggressively are on the rise.
The cumulative evidence suggests that, in practice, our tax system may be much less progressive than the statutory tax rates would suggest, which raises concerns about the possible corrosion of the tax system. In a self-assessment system such as New Zealand's, it is critical for taxpayers to accept that the tax rules are broadly fair and even-handed. There is a danger that the emerging pressures we have identified may undermine the willingness of individuals to comply voluntarily. This needs to be taken into account when considering future changes. A key consideration when examining future possible tax changes is whether they increase or decrease the robustness of the tax system.
International comparisons
The income splitting and tax sheltering opportunities we have highlighted are not unique to New Zealand and, at least on the surface, there would appear to be greater problems in many other countries that have greater differences between the top and the bottom personal marginal tax rates or between the top personal tax rate and the company rate.
In particular, New Zealand's gap between its company tax rate and top personal rate looks small when compared with Australia's (a six percentage point gap as opposed to Australia's 18.5 percentage point gap).[25] However, New Zealand may be less able to sustain a major difference because of the absence of a capital gains tax. The absence of a capital gains tax allows those on the top marginal tax rate to realise profits from a company as a tax-free capital gain rather than income. This can lead to the 33% company tax rate being very similar to a final tax when profits can be accumulated for a sufficiently long time before eventual distribution. We are not advocating the introduction of a capital gains tax. The question of whether to introduce a capital gains tax was examined by the Tax Review 2001, which concluded that "the disadvantages of taxing capital gains on a realisation basis outweighed any theoretical benefits from extending the base in this way". However, if the company tax were reduced significantly, relative to the top personal marginal rate, this issue might need to be reconsidered. For example, in Ireland a capital gains tax is considered essential to maintain the large difference between its company tax rate of 12.5%, and the top marginal personal income tax rate of 48%.[26]
Measures to deal with income splitting and tax sheltering
Perhaps because of greater differences in tax rates, other countries have done more than New Zealand in directly targeting income splitting and tax sheltering. Measures that deal with income splitting and sheltering are likely to be more administratively complex than aligning rates. Effectively, people's incentives to avoid tax increase the less rates are aligned, and preventing avoidance necessarily gets more difficult and necessitates more complex rules. Unless the government is willing to move to closer alignment of rates, we may need to consider some of these more complex rules.
When looking at countries with tax systems similar to New Zealand's (Australia, Canada, the UK, the US and Ireland), the most common measures have been:
- the alignment of trust and top personal tax rates
- income attribution rules
- rules to prevent the sheltering of personal services income through a company and
- excess retention taxes on the undistributed income of a company.
The most common approach taken in these countries has been the alignment of trustee and top personal rates, thus removing the incentive to use a trust to shelter income. In Australia, the UK, Canada and the US, the trustee rate is generally aligned with the top personal rate. Ireland also largely aligns rates by imposing a penalty tax on the undistributed income of discretionary and accumulation trusts, effectively equating the trustee rate with the top personal rate.
Income attribution rules are in place in Canada and the United Kingdom. These "attribution" (Canada) or "settlement" (UK) rules are aimed at preventing income splitting with a taxpayer (including via a trust) subject to a lower tax rate. Transferred income can be attributed to income of the person making the transfer. These rules can be both complex and costly to administer, and are subject to a number of exceptions limiting their effectiveness. More recent reforms attempting to plug these gaps in Canada have added another layer of complexity yet still result in exploitable gaps. Similar rules apply specifically to trusts in the US.
Rules are in place in Australia, the UK, the US and Canada to prevent the sheltering of personal services income through a company. These rules, generally termed "personal services company" rules, have been implemented in two ways. Company income can be attributed to the owner(s) of the company, as happens in Australia, the UK and the US. Alternatively, the company can be denied deductions and any other benefits that an individual would not be entitled to, as occurs in Canada.
Although other measures implemented overseas have had mixed success, there appears to be some consensus as to the benefit of personal service company rules. While also somewhat complex, these rules appear very successful in countering the sheltering of income from personal exertion in companies. New Zealand already has rules that cover the most blatant forms of sheltering personal exertion income. However, as long as the top personal and company rate are not aligned there appears to be a case for more comprehensive reform such as that in place in Australia.
The US and Ireland both charge a penalty tax on excess retentions of income to prevent sheltering within a company. In some cases, Canada imposes a tax on investment income to promote distribution. The tax is then refunded on distribution. These rules appear to be complex, and have potential repercussions on companies' incentives to reinvest retained earnings. Considering such measures requires some caution.
Possible directions for reform
There are a number of possible future directions for reform.
First, if there were sufficient fiscal headroom, one possibility would be to lower the top personal marginal tax rate to reduce discrepancies between the company tax rate and the top personal marginal rate. This option becomes more difficult if a substantial reduction in the company tax rate is being contemplated.
A second alternative is to increase the company and trustee tax rate to the top personal marginal tax rate. However, there are obvious problems with this direction of reform, given the increased incentives it would provide for transfer pricing and thin capitalisation. It is also likely to discourage inbound investment and hinder growth. Moreover, moves in this direction would not affect incentives for income splitting within a family, which largely depends on differences in rates of personal income tax.
A third option might involve some intermediate approach such as reducing the top personal marginal tax rate to 36%, increasing the trustee tax rate to 36% and reducing the company tax rate to 30%. This would maintain a six percentage point difference between the company tax rate and top personal rate and thereby not increase corporate tax sheltering opportunities for those on the top personal marginal rate relative to the status quo. At the same time it would remove the opportunity for trusts to be used to shelter income from the top personal marginal rate of tax.
The fourth possible direction of reform is to consider introducing some of the measures that other countries have adopted to directly confront income splitting and tax sheltering. This would, however, be likely to complicate the tax system. Even if there were to be greater tax rate alignment under one of the first three options, some move in this direction may be necessary. This is because tax sheltering schemes may be used not only to shelter income from higher rates of personal tax but also to shelter income from the very much higher effective marginal tax rates which arise when family assistance is being abated or child support is being assessed.
Key themes in future work
Base maintenance and remedial legislation will always be key features of the tax policy work programme.
In addition, a key theme over the next three years should be to consider measures to make the personal income tax system more robust. This could involve tax rate changes or strengthening anti-avoidance measures aimed at countering income splitting and tax sheltering - or a combination of the two. An important goal is to ensure that as much as possible of a person's tax liability is independent of how he or she structures the vehicle through which income is earned.
A second key theme is to increase the robustness of our tax rules on foreign investment in New Zealand. This theme stems from the importance of company tax as a revenue source for New Zealand. There are strong international pressures on the company tax base, and any major reduction in this base is likely to place upward pressure on other tax rates. One way of increasing robustness is to lower the company tax rate; another is to step up measures that counter tax avoidance and international arbitrage by companies operating in New Zealand.
When New Zealand introduced its full imputation system, in 1989, it decided that it would not provide imputation credits to foreign investors. The aim was to tax profits earned by foreign-owned firms that are sourced within New Zealand.
At present, however, some strong pressures are tending to erode source-basis taxation when companies are wholly or partly foreign owned. These include:
- streaming of imputation credits which would normally be received by foreign residents back to New Zealand residents and
- diversion of New Zealand profits to foreign countries and then streaming them back to foreign parties without full New Zealand tax being paid, as a result of our conduit or dividend withholding payment provisions.
Australia is a very important foreign investor into New Zealand, and when Australian firms invest into New Zealand through branches or subsidiaries the Australian full imputation system provides incentives for profits to be streamed to Australia to avoid New Zealand tax.
The incoming government will also need to consider other important international issues. For example, concerns have been expressed that New Zealand's relatively high rates of withholding tax on dividends, interest and royalties can provide an impediment to international investment. Australia has raised the issue of withholding tax in relation to our double tax agreement, to which we need to respond by the end of the year. There can be reasons to be cautious about reducing rates of withholding taxes. Withholding taxes on interest and royalties can be important in helping sustain source basis taxation, which is especially important if New Zealand wants to maintain a significant company tax base.
In our view, progressing both these international taxation concerns and measures to make the personal tax rate system more robust are key priorities for the next three years.
[15] This consideration led the Tax Review 2001 to suggest options for reducing the New Zealand tax rate on companies to 18% to the extent the company is owned by non-residents. These options were considered by the government but ultimately rejected on the grounds of fiscal costs and administrative impracticality.
[16] For evidence on transfer pricing and thin-capitalisation within OECD countries see Bartlesman and Beetsma, 2003, "Why pay more? Corporate tax avoidance through transfer pricing in OECD countries", Journal of Public Economics, 87, 2225-2252.
[17] A recent paper by Lee and Gordon (2005) suggests a strong negative relationship between the company tax rate and economic growth (see, Lee, Y., and R.H. Gordon (2005), "Tax Structure and Economic Growth", Journal of Public Economics, 89, 1027 - 1043.
[18] See James Kelly and Robert Graziani - "International trends in company tax rates - Implications for Australia's company income tax", and the OECD Tax Database. For countries with a progressive structure of company tax rates, the top marginal tax rate is recorded. Because these data do not include state taxes and surtaxes they will tend to understate total tax rates in some other countries. The data show the standard company tax rate and ignore the fact that some countries may have preferential tax rates for certain activities.
[19] The ten ASEAN countries are Brunei, Cambodia, Indonesia, Laos, Malaysia, Myanmar, the Philippines, Singapore, Thailand and Vietnam.
[20] There can also be non-tax reasons, such as asset protection, for using companies.
[21]Asset protection can also be a non-tax reason for using trusts.
[22] Up to $100,000.
[23]The figures for higher income individuals are less reliable because they are based on a sample of incomes, meaning the margin of error is higher for them because fewer taxpayers are involved. The year 1998-99 is used as the base year because businesses paid quite unusually high levels of dividends in 1999-00. This was likely to have been influenced by the increase in the top marginal tax rate from 33% to 39% in the following year.
[24]Some trust returns remain outstanding for 2003-04, so the increases are likely to be a little higher once all the returns are completed.
[25] Of course, Australia's larger gap may also be problematic.
[26] "All in rate" including social security and regional taxes.
Continues in "Key themes in future work"
Date published: 16 Nov 2005
Back to top