3. Policy challenges
In our view, an important policy challenge is ensuring a coherent ongoing tax policy framework. We have a relatively coherent system which makes it relatively difficult for people to escape their intended tax liabilities but maintaining coherence is a challenge.
Ours is not the only possible coherent tax system. Governments with different distributional objectives and different views on the importance of promoting economic efficiency and growth may reasonably come to different judgements about how progressive the tax system should be. There are coherent frameworks which can respond to different political judgements on this issue.
There are also coherent frameworks that tax capital and labour income in different ways that have been discussed in a number of recent reviews. In this chapter we examine what we see as three leading contenders for coherent structural approaches (or paradigms) for designing tax policy. There are pros and cons with each of them.
Key considerations include whether it is viable to continue with a broad alignment of tax rates and whether it is desirable to cut tax rates on investment and saving. Deciding on the over-arching structure of the tax system is the most fundamental tax policy decision a government can make and is a key tax policy choice for the Government. Until decisions are made on this matter, it is difficult to focus tax policy work on many important detailed design issues.
Given that New Zealand is starting from a much better place than most OECD countries, we think there should be a substantial burden of proof exercised before moving from our current broad-base, low-rate (BBLR) policy paradigm. Fundamental tax reform runs a risk of making things worse and less coherent. Moreover, given some features of the New Zealand economy, we believe that our BBLR structure is particularly well-suited to New Zealand.
Characteristics of a good tax system
Tax policies are generally judged against a series of sometimes conflicting criteria, that have been well-rehearsed in past tax reviews (see, for example, the report of the Victoria University Tax Working Group in New Zealand). These include:
Efficiency and growth: Taxes should be efficient and minimise impediments to economic growth.
Equity and fairness: The tax system should be fair. This involves both horizontal equity (fair treatment of those in similar circumstances) and vertical equity (fair treatment of those with differing abilities to pay tax).
Revenue integrity: The tax system should minimise opportunities for tax avoidance and arbitrage and provide a sustainable revenue base for the Government.
Fiscal adequacy: The Government should raise sufficient revenue to meet its requirements.
Compliance and administration costs: These should be kept to a minimum.
Coherence: Individual reform options should make sense in the context of the entire tax system. While a particular measure may seem sensible when viewed in isolation, implementing the proposal may not be desirable given the tax system as a whole.
Coherence is really a means of satisfying the other objectives outlined above rather than being an end in itself. The importance of the coherence criterion has been emphasised by the Mirrlees Review when it stated:
The tax and benefit system should have a coherent structure based on clearly defined economic principles. There should be a clear vision of the ideal system, in which the various elements fit properly together and from which unnecessary distortions have been eliminated.
A coherent tax system needs to fit together. For example, it is common internationally for governments to decide that personal income should be taxed at a set of increasing marginal tax rates. For such a system to be coherent it is vital that the statutory tax rates on personal income should “stick”. The tax system loses coherence if this progressive tax system can be circumvented by, for example, individuals sheltering income in trusts or companies. Similarly, the tax system loses coherence if there are arbitrary differences in the ways that different forms of savings or investment are taxed.
Any practical tax system is likely to lead to some cost in terms of reduced efficiency and growth because taxes affect whether and how hard people work, whether they train to increase their skills, whether they invest in risky but potentially profitable activities and how much they save and invest. The less coherent taxes are, the bigger these costs are likely to be by biasing people and firms to act in ways that would not be sensible in the absence of tax considerations. A lack of coherence also makes the tax system unfair because people in similar circumstances can pay very different amounts of tax and because tax breaks are often of most benefit to the well-advised and wealthy. A lack of coherence provides fertile ground for tax avoidance and arbitrage schemes. It makes the tax system a much less secure base for financing Government expenditure and tends to make the tax system more costly to comply with and administer than would otherwise be the case.
New Zealand’s tax system must be broadly understood and seen as reasonable by the New Zealand public. If it is not, it will not be sustainable and generally will inevitably be changed. A coherent tax system helps ensure that the tax system is understood and viewed as reasonable.
Deviations from coherence which at first may seem small can matter. The adverse impacts of a non-aligned set of tax rates were documented in Inland Revenue’s BIM in 2008 and were also a key concern of the recent Victoria University of Wellington Tax Working Group. This lack of coherence has been largely removed through the recent alignment of the top personal tax rate with the trustee tax rate and the reduction in the gap between the top personal rate and the company rate.
We believe that New Zealand’s BBLR tax system is one of the most coherent in the OECD. There are however coherent alternatives that have been implemented in other countries (notably Norway) or suggested in overseas tax reviews. These differ in the extent to which they tax residents on their capital relative to their labour income and how heavily they tax non-residents on their investment into a country. These issues have figured prominently in reviews of taxation in New Zealand and in other countries (including Australia and the United Kingdom).
A key goal of successive governments has been productivity. Productivity requires that capital and labour are put to their best uses and this is best served by reducing, as much as possible, biases in the way that different forms of capital and different forms of labour are taxed. We believe that taxing different activities as neutrally as possible within some coherent framework is likely to be critical for efficiency and productivity and probably more important than the choice between coherent frameworks.
Coherent tax policy options
Coherent tax policy options include:
- a broad-base, low-rate (BBLR), broadly aligned system (the New Zealand paradigm);
- a Nordic tax system; and
- the Mirrlees approach.
Broad-base, low-rate broadly aligned system (the New Zealand paradigm)
The New Zealand system applies a broad consumption tax across as many forms of consumption as possible and a broad income tax over most types of income. By keeping tax bases broad, the GST and income tax rates are kept relatively low. This minimises economic distortions.
Arguments about the breadth of the GST base are largely independent of choices between the three coherent tax schemes outlined above and the breadth of the GST base is not discussed further in this BIM.
Under New Zealand’s BBLR framework, income from both capital and labour is taxed at progressive marginal rates aimed at providing a fair way of taxing those with different capacities to pay tax. As much as possible, this scheme is aimed at ensuring that all income (both capital and labour income) is taxed at the marginal tax rates of New Zealanders.
An important issue is how to apply a consistent framework to income earned through entities such as trusts and companies.
In New Zealand, income earned through trusts can be retained within a trust and taxed as “trustee income”. This is a final tax. However, scope for tax sheltering has been removed for most taxpayers by aligning the trustee tax rate with the top personal tax rate.
Income earned within companies is taxed at the company rate, subject to a full imputation system. When company profits are distributed to shareholders as dividends, imputation credits are provided for company tax that has been paid. Company profits end up being taxed at the marginal rates of shareholders. If the company tax rate were too far below the top personal rate, this system would break down by allowing personal income to be sheltered within companies. Even though income would eventually be taxed at shareholders’ rates upon distribution, there would be important timing benefits. But at current tax rates, where the gap between the company rate and top personal rate is only 5 percentage points, the scope for sheltering is limited.
The company tax rate also provides a final tax on non-residents investing into New Zealand through companies.
The New Zealand system achieves coherence very simply. Even within its BBLR paradigm, it has been possible to alter the balance in favour of taxing capital less heavily and labour more heavily by increasing the rate of GST and lowering income tax rates. A GST taxes labour income. It reduces the real goods and services that can be purchased out of the income earned by providing an extra hour of work just like a labour income tax would. Of course there are both efficiency and equity issues that need to be considered when undertaking this sort of tax rebalancing. Similar issues need to be thought through when there is any rebalancing which reduces tax on capital income and increases tax on labour income.
It should be acknowledged that New Zealand’s tax system falls short of ensuring that all income is taxed at personal tax rates in a fully consistent fashion. The fact that the company tax rate is less than the top personal rate provides some timing benefits when income is accumulated in companies. The portfolio investment entity (PIE) system also provides some tax advantages. For example, the capped PIE rate means that individuals on the top marginal tax rate are taxed at 28% rather than their higher rate of 33% on income accumulating in PIEs. But by OECD standards, these are relatively small inconsistencies.
Nordic tax system
Continental European countries generally have high ratios of tax to GDP and wish to subject labour income to relatively high levels of taxation, especially when social security taxes are combined with personal income taxes. Given the mobility of capital, especially between countries in close economic and geographic proximity, such high rates of tax applied to capital income have proven unsustainable. Accordingly, most EU countries have introduced some form of reduced taxation for capital income.
The Nordic countries, especially Norway, have attempted to do this in a coherent manner by explicitly taxing all forms of capital income at a lower rate than labour income. Capital income includes forms of income which are a return on invested capital including interest, dividends, rental income and the return on capital invested in a business. Labour income is income from personal effort including salaries and wages as well as the returns from the owner of a closely held business working in that business.
Generally the system applies a lower rate of tax on income from capital than is applied to labour income. The system therefore requires rules to distinguish capital and labour income. Distinguishing capital from labour income has proven difficult to achieve in practice.
Initially Norway treated widely held and closely held companies differently with all income in widely held companies treated as capital income while income from closely held companies was split into labour and capital components. But this became problematic as it was much more advantageous to be treated as “widely held”.
In response to these problems, Norway recently revised its system to treat all companies alike. Companies are taxed at the capital tax rate. Shareholders receive returns (by way of dividends or capital gains) below a “normal” or “risk-free” return without any additional shareholder tax. Returns above this are taxed in shareholders’ hands at the capital tax rate. This means that the amount that approximates labour income ends up being double taxed at the capital tax rate. This approximates tax at the top rate of labour income tax (given the particular tax rates chosen by Norway). This is a clever method of ensuring that companies cannot be used to shelter labour income from high rates of labour income tax. Capital and labour income must still be separated for unincorporated businesses.
Non-residents are taxed on capital income derived through Norwegian companies at the capital tax rate of 28%.
Mirrlees Review proposal
The Mirrlees Review has proposed major revisions to the tax system in the United Kingdom. It has proposed that taxpayers be able to claim a deduction for a risk-free return on most forms of savings. This is referred to as a rate of return allowance (RRA). This ends up having similar effects to completely exempting capital income from tax. Thus, the Mirrlees Review proposal can be thought of as equivalent to an extreme Nordic tax system with a zero tax rate on capital income.
The aim of effectively exempting capital income from tax is to reduce taxes on savings and to reduce biases between different forms of saving.
Businesses would face a substantially lower average rate of tax as they would only be taxed on amounts in excess of the risk-free return. This would be achieved by providing an allowance on corporate equity (ACE) equivalent to the risk-free return. Income in excess of the risk-free return would be subject to normal tax rates. The effect of these systems would be to eliminate taxation for marginal investments, while continuing to impose tax on economic rents (that is, returns over and above those required to induce investment).
Reasons for choosing between the three options
In choosing among the three options, two fundamental design issues are raised:
- How heavily should New Zealand be taxing non-residents on their investments into New Zealand?
- How heavily should New Zealand residents be taxed on their capital income relative to their labour income?
Taxing non-resident investment into New Zealand
One of the key features of company taxation is that it taxes non-residents investing into New Zealand. A potential concern with company taxation is that it can discourage investment into New Zealand. In small open economies like New Zealand, foreign investors will demand comparable returns to those that can be obtained from investing elsewhere. For example, if foreign investors can earn 6% after-tax from their investment in other countries, they will not accept less from investing into New Zealand. A company tax rate of 28% would drive up the required pre-tax rate of return on investments in New Zealand to 8.3%. Thus, company taxation will tend to drive up the before-tax returns that companies need to earn in order to provide an attractive after-tax return to foreign investors. This reduces investment and New Zealand’s capital stock. With fewer computers, tractors, factories and so forth, New Zealand labour becomes less productive than it would otherwise be and workers earn lower wages. Thus, high taxes on inbound investment generated by a high rate of company tax can end up hurting New Zealanders. Moreover, standard economic theory suggests that these taxes can be more harmful to workers than would be taxes on their labour income that collected the same amount of revenue. By itself, this suggests a theoretical argument for eliminating company tax altogether.
However, some firms investing into New Zealand may be earning “economic rents” associated with firms locating themselves in New Zealand. These are returns over and above those required for them to invest in New Zealand. These economic rents can arise because many non-resident firms invest into New Zealand in order to sell to the domestic market, (rather than export into world markets). This can generate returns significantly in excess of those that the foreign firm demands in order to invest. Taxing these economic rents at the company rate of 28% may still provide the foreign firm with an after-tax return above that which they would require to invest into New Zealand. In these cases the company tax rate of 28% (or even a significantly higher rate) will be borne by the foreign firm. To reduce the rate to zero would provide a windfall gain to foreign shareholders and may do little to affect foreign investment in New Zealand. To the extent that reducing the company rate reduces Government revenue, this is likely to put upward pressure on other tax rates. This has the potential to make New Zealanders as a whole worse off.
New Zealand residents – how heavily to tax capital relative to labour income
The second issue is how heavily should New Zealand residents be taxed on their capital relative to their labour income. There are two key arguments that can be put forward for lower tax rates on capital income. First, any tax on capital income lowers after-tax returns and tends to provide a tax bias discouraging savings. There are economic models where any tax on capital income is a “bad thing” but other economic models where some taxes on capital income are desirable. While determining the most efficient tax rate on capital income is far from being resolved, there is reasonably common agreement amongst economists that high taxes on savings may be inefficient. A second reason for cutting tax rates on capital income is to reduce tax biases between different forms of savings (such as whether one saves in a traditional interest-bearing bank account where the full nominal interest is taxed, or perhaps through acquiring an owner-occupied house or a rental property). Clearly, the lower the capital tax rate the lower will be the tax distortions between different forms of saving.
Effects of the different options
If these were the only considerations, there would be pros and cons in a shift from our BBLR system in a Nordic direction. If the tax rate on capital were to fall, it would reduce a tax bias against inbound investment and reduce disincentives to save. At the same time, it would reduce our ability to tax economic rents.
On the surface, a shift in the direction of the Mirrlees Review proposals would seem preferable. This would, in theory, eliminate any bias against investment into New Zealand and also any bias against saving. At the same time it would continue to allow us to tax economic rents. But the Mirrlees Review suggestions are largely untested. There are likely to be large costs in a small country like New Zealand being the first to implement such a new system.
A shift in either the Nordic or Mirrlees Review directions would reduce revenue from taxes on capital income. As the Mirrlees Review suggestion is effectively to eliminate capital taxation, the revenue reduction in this case would be large. Thus, merely looking at how these various options alone affect inbound investment or savings is very partial. If other taxes are required to replace the forgone revenue, the relevant question is how distorting these replacement taxes are likely to be relative to the taxes they are replacing. The main options would appear to be higher rates of tax on income (or labour income with a Nordic tax system) or a higher rate of GST. These taxes will create their own inefficiencies. For example, high rates of labour income tax can encourage talented people to leave New Zealand or discourage people from taking risks or acquiring new skills. It encourages people to work in untaxed ways at home rather than in ways that produce taxable income. This can lower productivity. It is important to note that it is not only taxes on capital income but also taxes on labour income that can lower productivity.
A key consideration is how large the benefits of a move in a radical new direction are likely to be. There is considerable uncertainty surrounding this issue but a study by de Mooij and Devereux (2009) quoted by the Mirrlees Review suggested that a shift to an ACE system in the UK and increasing value added taxes (GST) to replace the revenue forgone would boost long-run investment by 6.1 percent, wages by 1.7 percent, employment by 0.2 percent, GDP by 1.4 percent and the welfare of a typical consumer by 0.2 percent of GDP. Welfare gains can be much lower than any increase in GDP because much of any additional GDP accrues to foreigners and extra work or extra savings will have an opportunity cost.
When considering the pros and cons of a switch to a radical new tax system, it is important to note that much of the devil is in the detail of possible reform options.
A case in point is Norway’s new method for taxing companies. The aim is to double tax labour but not capital income earned through companies. In practice, anti-avoidance provisions have meant that there can be considerable double taxation of risky income earned through companies. This can penalise those who invest in risky but potentially high-return investments.
The Mirrlees Review proposals are harder to grapple with because they are largely untested although Belgium has introduced an ACE company tax system. We expect there would be many difficult issues to work through. In particular, it may be hard to define equity satisfactorily and to prevent international tax avoidance with such a scheme. We understand that there have been considerable tax avoidance pressures in Belgium. Scope for international tax arbitrage can occur when a country adopts a different conceptual basis for taxation from other countries. Such arbitrage can give rise to unexpected losses of tax revenues.
Features unique to New Zealand
New Zealand has the following unique features that influence its tax policy environment relative to other countries:
- It is geographically isolated from major markets – this implies that it is less likely to attract export industries by lowering its company tax rate on non-residents.
- It is likely that FDI into New Zealand may be less sensitive to tax given that we are a long way from world markets and hence New Zealand is less likely to be a venue for a global business. Instead, most FDI into New Zealand may be directed at businesses serving the domestic market.
- It is likely to have much more location-specific economic rents than would be true of less isolated countries. (Foreign firms operating in New Zealand tend to sell to the local market which makes economic rents less mobile and more easily taxed. By contrast, a small landlocked country in Europe may have limited scope for taxing economic rents because if it tries to do so, firms can relocate and supply much the same market from just outside the border.)
- It has a high mobility of labour. This implies that New Zealand may suffer disproportionately from having high personal tax rates applied to labour income.
- The structure of the tax system (the imputation system and the low top personal rate) makes simplification through rate alignment feasible in New Zealand, while in other countries the benefits from rate alignment are simply not practically attainable.
- New Zealand has a singularly broad and efficient GST. In the past this has made increasing the rate of GST and lowering income taxes a more attractive way of switching from taxing capital income to taxing labour income than it would be in most other OECD countries. However, increasing the rate of GST further is likely to make it more difficult to maintain a relatively broad GST base.
These features of the New Zealand tax system all make a BBLR approach for New Zealand more attractive than it might be for some other countries.
Structural implications of the three approaches
The three alternative tax systems have different structural implications:
- Under New Zealand’s BBLR system, the trust tax rate needs to equal the top personal tax rate. If these rates are allowed to diverge, (as in the 2000 to 2010 New Zealand system), the system loses coherence and is vulnerable to tax avoidance activity as documented in our previous BIM. In addition, the BBLR works best if (as at present) the company rate and top personal rate are not too far apart. Countering tax avoidance when tax rates diverge markedly requires the introduction of complex anti-avoidance rules to prevent deferral or avoidance of tax for certain types of personal income earned through companies.
- Under the Nordic system, there is scope for differentiation of tax rates on capital and labour. Norway’s design results in a company tax rate that is approximately half the top tax rate on labour income. However, Norway’s system relies on maintenance of a fixed relative set of rates and needs a distinction between capital and labour income to be implemented for unincorporated businesses. Moreover, coherence requires tax on capital gains.
- The system proposed in the Mirrlees Review allows for a zero effective tax impost on marginal investments while still taxing economic rents. A key question is the viability of making this tax system work.
Policy issues with the approaches
Important policy issues underlie the choice for New Zealand among the approaches. These include:
- the viability of continuing with a broad alignment of tax rates; and
- whether it is desirable to cut tax rates on investment savings.
1. Alignment of tax rates
The major question concerning the longer-term viability of the New Zealand tax policy paradigm is whether international pressures will force a wider divergence between the company and top personal tax rate.
Among OECD countries there has been a continuing trend toward lower company tax rates, although the aggregate share of company tax as a percentage of total tax collected has not fallen worldwide. Even with the recent reductions in the company tax rate, New Zealand has moved from having one of the lowest company tax rates in the world to one that is above the average tax rate internationally.
The key questions are whether these factors will drive New Zealand to further reduce its company tax rate and, in particular, whether any patch-ups to make a less aligned system work would become so cumbersome and unattractive that it would be sensible to abandon the BBLR framework.
Setting the company tax rate
The company tax rate has a dual role in supporting domestic tax settings on individuals while taxing non-residents on income sourced from New Zealand. Tradeoffs are involved in setting the company tax rate. As discussed earlier, too high a rate could discourage efficient investment. It also increases tax biases between different corporate investment options and creates transfer pricing pressures.
Too low a rate fails to tax immobile economic rents adequately and increases pressures from arrangements designed to shelter income from personal tax.
As noted above, New Zealand’s particular situation appears to render reductions in the company tax rate less beneficial in attracting investment than they might be in other jurisdictions. Reducing the company tax rate may increase investment by making it more attractive for firms producing for a regional or world market to locate in one country in the region rather than another. But being a long way from other markets is likely to make investment less sensitive to the company rate. It reduces the numbers of firms that are likely to locate themselves for reasons other than supplying the domestic market. At the same time there are probably higher costs than for other less isolated countries through forgoing tax on immobile economic rents. This suggests that New Zealand should not be leading the international charge in moves to lower its company tax rate. At the same time, transfer-pricing and thin-capitalisation pressures are likely to mean that New Zealand’s company rate cannot get too far out of line with other countries.
Whether recent trends in cutting company tax rates will continue is an open issue. OECD countries face significant pressures to maintain and increase revenues following the global recession and in anticipation of significant demographic-based pressures on health and pension costs. Most of the possible base-broadening has already been squeezed out of corporate tax systems, so that cuts in the company tax rates would need to be compensated for out of taxes that more obviously bear upon individuals. Whether this is feasible in an era of continuing spending pressures is questionable.
Setting the personal tax rate
In addition, New Zealand faces pressures to keep the top personal tax rates lower than other countries. Setting the top personal tax rate involves a trade-off between fairness and efficiency objectives. A higher top rate applied to higher income taxpayers may be seen as desirable by a Government keen to fund transfers to lower-income taxpayers. On the other hand, higher tax rates reduce work effort and the incentive to invest and save.
New Zealand has to be concerned about taxes on labour because of its high mobility of labour. An important factor in boosting productivity in New Zealand is likely to be ensuring that it is attractive for New Zealanders to upskill, take risks and for highly productive individuals to remain in the country. Higher marginal tax rates on labour income would reduce these incentives.
Alignment need not be exact to have a stable system. It is critical to align the top personal tax rate and the trust tax rate as the trust tax rate is a final rate. This means that any tax benefit achieved is a permanent benefit. The compounding of this benefit over time means that a substantial after-tax gap can emerge between the returns on an investment made through a trust and the same investment held directly by an individual.
A moderate gap between the top personal tax rate and the company rate is sustainable due to New Zealand’s imputation system. Imputation ensures that the total amount of tax paid on income earned through a company and distributed as a dividend is the same as that paid if the income were earned directly. Differences are timing, not permanent benefits. When the gap between the company tax rate and personal tax rates is not too large these benefits are insufficient to affect behaviour. On the other hand, if the gap is too large, complex mechanisms are required to prevent recharacterisation of income. Taxation of capital gains on shares would also be necessary in such a case to prevent recharacterisation of dividends as sales of shares giving rise to capital gains.
This means that New Zealand’s broadly aligned BBLR approach remains viable under current tax settings. However, complex patch-ups would be necessary and ultimately its viability would be called into question if there were a desire to either lower the company tax rate substantially or to increase personal rates of tax substantially.
2. Incentives to save and invest
Moving in the direction of either a Nordic tax system or the Mirrlees Review proposal would boost incentives to save and invest. At the same time replacement taxes are likely to increase biases in other areas.
Key downsides a Government would need to address are the perceived fairness of giving large tax cuts to those with high levels of capital assets and large amounts of capital income, and how to replace the forgone personal and corporate tax revenue. Would the Government be willing, for example, to reduce spending or boost income tax rates on labour income or the rate of GST in order to introduce an ACE and RRA system? If the Government is not willing to raise tax rates elsewhere or do other things to bridge the fiscal gap, the benefits of any moves in this direction are largely academic. This is perhaps the key question that a Government considering fundamental reforms would need to address.
While there is no “ideal” tax system, we believe that New Zealand’s BBLR broad-alignment paradigm continues to be viable and remains the best path for New Zealand to follow at present. A critical issue for the New Zealand tax system over the past decade has been coherence. Large economic distortions are likely to arise when tax systems are not coherent. Misalignment of rates has led to significant levels of tax avoidance activity. Looking around the world, it is very hard to be confident that replacing New Zealand’s current largely coherent tax system and bringing in an alternative would improve the performance of the tax system. Indeed, there is a substantial risk it would perform more poorly as a stable source of revenue that has broad public acceptance. The fact that we have a tax system that works well suggests a high onus of proof in altering our basic tax settings.
A key issue is that New Zealand’s tax system must be broadly understood and seen as reasonable by the New Zealand public. If it is not, it will not be sustainable and it will inevitably be changed. Unsustainable tax reform is the worst kind of tax reform.
Perhaps the most fundamental tax policy question you need to consider is whether at least over the next three-year period you wish to retain New Zealand’s BBLR broad-alignment paradigm.
It is important to confirm whether or not the Government is continuing with the current policy settings or wishes to redesign the tax system in a fundamental way. Resolving this matter would allow businesses to plan their affairs with more certainty about future tax implications, and tax policy resources would be freed to work on addressing concerns that have arisen within this basic framework.
14 A GST also taxes the wealth that people have on the day the tax is introduced by reducing its purchasing power. A GST is sometimes characterised as a tax on labour income combined with a lump-sum tax on wealth.