Australia's Future Tax System

Final Report: Detailed Analysis

Chapter B: Investment and entity taxation

B2. The treatment of business entities and their owners

B2–4 Reform directions for dividend imputation

Recommendation 37:

Dividend imputation should be retained in the short to medium term, but for the longer term, consideration should be given to alternatives as part of a further consideration of company income tax arrangements.

Recommendation 38:

A flow-through entity regime for closely held companies and fixed trusts should not be adopted for now, but would merit further consideration if there is a move away from dividend imputation in the long run.

Recommendation 39:

While dividend imputation is retained, imputation credits should continue to be provided only for Australian company income tax. Dividend streaming and franking credit trading practices should, in general, continue to be prohibited.

Recommendation 40:

If increased integration of the Australian and New Zealand economies is desired, a broad examination of the appropriate degree of harmonisation of business income tax arrangements between Australia and New Zealand should be undertaken.

Alternatives to dividend imputation should be considered for the future

Dividend imputation continues to deliver benefits for Australia, particularly for smaller firms and those operating in the more closed segments of the economy. However, a continuation of the trend of increased openness, rapid growth in cross-border investment flows and greater capital mobility will reduce the benefits of imputation in the longer term. Consideration therefore needs to be given to long-term reform options that provide a better fit with the global economy but which still retain the positive aspects of imputation (see Recommendation 37).

For a small, open economy that is increasingly integrated with international capital markets, providing tax relief only on dividends paid to resident shareholders will become less effective in reducing the cost of capital for companies (and hence of reduced benefit in encouraging investment) or in providing a neutral treatment of debt and equity. The bias for domestic savings to be invested in the shares of Australian companies will increase, limiting opportunities and increasing risk to households from poorly diversified savings portfolios.

Reform could involve switching tax relief from the double taxation of dividends from the shareholder level to the company level. Doing so would provide the same outcomes for resident shareholders as the current system, but would further reduce tax on non-resident shareholders. In effect, there would be a move to greater reliance on the taxation of residents' savings income and less reliance on source-based investment taxes. This switch would further encourage investment in Australia and reduce Australian companies' reliance on foreign debt. It would also reduce biases in the allocation of residents' savings.

To achieve such a switch, consideration could be given to a partial integration system that is common overseas, while at the same time reducing the company income tax rate. A more radical approach, which has greater potential gains but is largely untested overseas, would be a move towards a company or business level expenditure tax. This option for long-term reform is outlined in Section B1 Company and other investment taxes.

A flow-through entity regime for closely held businesses

The Australian Government asked the Review to consider a proposal to allow small, closely-held companies and fixed trusts the option to effectively be treated as partnerships for tax purposes. Under this approach, income and losses of the company or trust would be assigned to shareholders and beneficiaries regardless of whether they were distributed. The proposal received mixed support in submissions.

The proposal has the potential to reduce the compliance burden for micro-enterprises, as the many sets of rules associated with the current separate entity treatment of companies and some elements of the treatment of trusts would not apply. For example, flow-through taxation would make redundant the deemed dividend rules relating to non-commercial loans from a company to shareholders. The proposal could also allow some multiple entity structures to be simplified.

Flow-through would also allow the tax losses of an entity to be transferred to its owners, who could then offset the losses against other income, rather than leaving the losses trapped in the company or trust. A flow-through regime could therefore also have the benefit of improving loss symmetry, a potentially useful policy outcome if measured tax losses correspond to economic losses (see Section B1).

While flow-through approaches to the taxation of business entities have general merit, flow-through entities could become yet another option for business to consider or another component of an even more complicated business structure. New rules would be required to determine eligibility for, and the consequences of, flow-through treatment, and transitions into and out of such arrangements. Where flow-through treatment is provided for businesses falling below a size threshold, the prospect of losing flow-through treatment could deter small businesses from expanding.

Experience with optional regimes suggests that they can significantly complicate the tax system while doing little to reduce compliance costs (see Section G5 Monitoring and reporting on the system). Research in the United States, where a number of company or company-like flow-through entities are available, has found that the income tax compliance costs of operating a flow-through vehicle are marginally greater than the costs for a normal company (under a classical company income tax) and around one-and-a-half times the costs of a general partnership (DeLuca et al. 2005).

While flow-through companies and related entities are extensively used in the United States, they were developed in the context of a system that at the time provided no credit at the shareholder level for company income tax paid. In Australia, dividend imputation provides reasonably effective integration between shareholders and companies, so the case for running multiple systems is weaker.

However, as part of any consideration of a long-term move away from dividend imputation, adoption of flow-through company and entity arrangements may be a useful means to provide appropriate outcomes for smaller businesses (see Recommendation 38).

Dividend imputation should retain its current features

So long as dividend imputation is retained, pressures on how it operates will continue to increase as the economy becomes more open and cross-border investment flows grow. In this regard, and generally reflecting the long-standing concerns of the business community, submissions to the Review have proposed altering the current imputation system to provide increased recognition for foreign taxes or to permit dividend streaming.

The primary argument for these proposals is to reduce the imputation bias against offshore investment. However, as discussed, to the extent that there is a bias, it may be beneficial to Australia. The case for change, therefore, is problematic in general and current policy settings should remain in place (see Recommendation 39). In addition, the proposals raise further specific issues that are discussed below.

Providing a tax credit for foreign tax is problematic

Submissions to the Review have proposed providing a credit for foreign investment taxes paid by Australian companies on the basis that this will remove impediments to Australian companies expanding overseas. Credits for foreign taxes could take the form of either an imputation credit for actual foreign company tax paid or a uniform credit that is not linked to actual foreign tax payments.

Since companies seeking to expand offshore would typically be larger and more mature, they should have better access to international capital than other businesses in the domestic economy. Providing imputation credits to resident shareholders for foreign tax paid would not directly assist them in raising foreign capital and so could have limited impact on their cost of capital and their potential for offshore expansion. It would, however, increase resident shareholders' post tax returns from their savings.

This argument may not apply in all cases. For example, there may be small, newly formed internationally or regionally focused businesses that look to expand offshore relatively soon after starting up. If such companies have only a few owners, there could be an incentive for the owners to relocate both themselves and their companies offshore. This is because a change of residence could reduce the total Australian and foreign tax bill at the company and shareholder levels. Providing increased recognition of foreign company taxes could go some way to reducing or reversing these incentives.

Providing a credit for foreign taxes paid by an Australian multinational could also have the benefit of removing a bias that encourages households and superannuation funds to invest more in domestically-orientated Australian companies than they otherwise would. At the same time, for investors wanting greater exposure to investments in foreign countries, it would create a bias favouring investments in Australian multinationals over foreign companies.

However, providing credits for foreign taxes would reduce the integrity benefits of the imputation system. By extending imputation to foreign taxes, the incentive for Australian companies to pay Australian tax would be reduced. Similarly, domestically owned companies would have greater incentives to shift profits to low-tax foreign countries, putting additional pressure on rules to prevent international profit shifting.

Providing a credit is difficult in practice

If imputation credits were provided to resident shareholders for actual foreign company income and withholding taxes paid by Australian companies, there would be increased administration and compliance costs associated with identifying creditable foreign taxes. Companies would need to identify and potentially track foreign taxes paid by foreign subsidiaries. The Australian Taxation Office would also need to be able to verify those payments, which would be difficult.

Given these practical difficulties, imputation credits provided for foreign taxes would not be refundable. Companies and shareholders would be required to account separately for refundable and non-refundable imputation credits, further increasing complexity and compliance costs.

A potentially simpler alternative would be to provide a non-refundable tax credit for dividends paid to resident shareholders by Australian companies out of designated foreign income not subject to Australian company income tax. Companies already track foreign income under the conduit foreign income rules. However, in this case credits could be provided to resident shareholders for dividends paid out of foreign income not subject to foreign tax.

Tax harmonisation with New Zealand and the mutual recognition of imputation credits

The merits of Australia and New Zealand recognising each other's imputation credits has been the subject of previous consideration by both governments, and has been proposed in submissions to the Review.

Mutual recognition of imputation credits would involve providing imputation credits for foreign taxes on a reciprocal rather than unilateral basis. Australian shareholders of Australian and New Zealand companies investing in New Zealand would be eligible to receive a credit for New Zealand company income tax paid. A similar arrangement would apply for New Zealand shareholders.

Mutual recognition would have the potential to improve the allocation of investments between the two countries, increasing productivity, and potentially reducing barriers to competition between Australian and New Zealand companies. It could also reduce incentives for firms to engage in profit shifting between Australia and New Zealand, probably to New Zealand's net benefit.

Some of the issues raised by bilateral mutual recognition are similar to those that would arise if Australia were unilaterally to increase recognition for foreign taxes. To the extent that the cost of capital for firms is set internationally, the benefits and costs of bilateral mutual recognition in respect of investment allocation would be reduced. Mutual recognition would also entail additional complexity and administration and compliance costs, though tax administration issues would be more manageable.

There is currently a significant imbalance in trans-Tasman investment. New Zealand direct and portfolio investments in Australia, as at 31 December 2008, totalled $14.3 billion ($4.5 billion and $9.8 billion respectively). In contrast, Australian investment in New Zealand totalled $36.2 billion ($32.5 billion and $3.7 billion respectively) (ABS 2008a). Hence, mutual recognition would impose higher revenue costs on Australia than on New Zealand. However, any imbalance in direct revenue costs is not of itself an argument against mutual recognition. Revenue costs do not necessarily represent transfers from Australia to New Zealand. They can also reflect reduced taxes imposed by Australia on the savings of its own residents (and vice versa).

The case for mutual recognition has also been raised in the context of developing closer economic relations between Australia and New Zealand. This is because mutual recognition could have a role in furthering broader policy objectives and achieving greater integration of the two economies.

If increased integration of the Australian and New Zealand economies is desired, the starting point for an assessment of the issues and possible benefits should be broader than just mutual recognition. Consideration would better start with a broader assessment of the appropriate degree of harmonisation of the two countries' business income tax arrangements (see Recommendation 40). That broader assessment would also take into account company income tax settings and related arrangements such as capital gains tax, the treatment of foreign source income, and the taxation of financial arrangements.

Mutual recognition would be one element of this broader examination, which could also take account of possible long-term reform directions in Australia.

Dividend streaming of foreign source income

The current rules prevent dividend streaming; that is, the payment of franked dividends to those shareholders who benefit from them the most (typically residents) and unfranked dividends to other shareholders (typically non-residents). Submissions to the Review have proposed allowing the streaming of unfranked dividends, paid out of a company's foreign source income, to non-resident shareholders. However, there are good reasons to prevent dividend streaming.

There are a number of variants of dividend streaming, but all would have the potential effect that a shareholder's tax interest in company profits would be different to their legal and economic interest. This would be inconsistent with the integration objectives of imputation. For tax purposes, non-resident shareholders would be assumed to have an interest in the company's foreign income ahead of its other income; whereas their economic and legal interest would typically be in both the domestic and foreign source income of the company (see Chart B2–1). A similar divergence would arise for resident shareholders.

Chart B2–1: Distribution of company profits under dividend streaming

Panel A: Current tax and economic position

Chart B2–1: Distribution of company profits under dividend streaming - Panel A: Current tax and economic position

Panel B: Possible tax position under dividend streaming

Chart B2–1: Distribution of company profits under dividend streaming - Panel B: Possible tax position under dividend streaming

Dividend streaming could reduce the tax burdens faced by resident or non-resident shareholders, but for those benefits to be realised, the company would need a mix of resident and non-resident shareholders and of (taxed) domestic source and (untaxed) foreign source income. To maximise the benefits of dividend streaming, companies would need to get the right proportions of resident and non-resident shareholders, and of domestic and foreign income.

Hence a small, rapidly expanding and domestically owned company seeking to invest offshore would not benefit unless it changed its ownership structure. Those companies best able to benefit from streaming — those that have a mix of resident and non-resident shareholders and investments — would be likely to have access to international capital and be less reliant on domestic financing. Hence, their cost of capital would more likely be set internationally, reducing the potential benefits from streaming.

By maximising, for any given amount of company income tax paid, the credits available to frank dividends to resident shareholders, dividend streaming would also reduce the incentives to pay company income tax. It would therefore reduce the integrity benefits of the current rules.

Dividend streaming by foreign multinationals

Another possible variant of dividend streaming would be to permit foreign multinationals with a secondary listing on the Australian stock exchange to stream the imputation credits from their Australian subsidiary to the resident shareholders in the parent company.

The benefits of doing so could be to encourage secondary listing on the Australian stock market by firms with significant operations in Australia, increasing demand for Australian financial services. Further, by increasing the degree of connection of foreign companies with Australia it could conceivably reduce any bias against investing in Australia stemming from it geographic isolation.

However, it is not clear that permitting dividend streaming in this way would have these effects, or that the tax-driven stock market activity generated would be worthwhile. In addition, under the current imputation system, a multinational with Australian shareholders and operations has an incentive to take up Australian (or New Zealand) residency. Permitting dividend streaming by foreign multinationals would remove this incentive.

The current imputation system encourages residents to over-allocate savings into domestically focused Australian companies, which may limit the scope for geographic risk diversification. Permitting streaming by foreign multinationals could offset this bias by encouraging Australians to own shares in these companies. However, in turn, the bias against investing savings in internationally focused companies (both Australian and foreign) would worsen.