Working paper summary

Dividend imputation or low company tax?

Geoffrey Kingston - Macquarie University

Data from the Organisation for Economic Cooperation and Development (OECD) offers limited support for the proposition that Australia’s company tax rate could be cut substantially with little or no loss of tax revenue.

However, analysis of the type conducted by the Federal Treasury indicates otherwise. This suggests that our headline company tax rate could be cut to 20 per cent if dividend imputation were abolished.

Dividend distributions from Australian companies that have paid company tax on their profits are termed franked dividends. Australian households, superannuation funds and charities that receive franked dividends are reimbursed for the tax paid by the company via the system for taxing individual and superannuation incomes. The total value of these reimbursements currently amounts to approximately $19 billion per annum. Furthermore, investors may claim franking credits even if they pay no tax, a facility that has become popular with self-funded retirees.

The question arises as to whether franking credits for households, superannuation funds and charities should be abolished, and the savings applied to cutting the headline rate of company tax.

It is hard to estimate with any precision the amount by which the headline rate of company tax could be cut. The Federal Treasury’s 2015 Re:think Tax Discussion Paper floats this idea but does not provide specific details.

The standard analysis of company tax in a small open economy has been reprised in the Treasury Paper. The analysis assumes that: capital is internationally mobile; labour is immobile; payout ratios are 100 per cent; corporate borrowing is either zero or unresponsive to tax policy; and the marginal investor resides offshore. Under these conditions, the size of the domestic capital stock is driven by the headline company tax rate and is unaffected by dividend imputation. This is the essence of the case for abolishing dividend imputation and using the proceeds to lower the headline corporate tax rate.

Part of the case for replacing dividend imputation with a lower headline corporate tax rate is that Australia has fallen out of step with its traditional peer group. Notably, Canada and the United Kingdom have lower headline tax rates than ours, currently 26 per cent and 21 per cent respectively. Moreover, Canada offers a dividend tax credit to shareholders in respect of domestic corporate profits, but does not appear to have the cash-refund feature that we have had since 2000. Full imputation is not the only quirk in our tax system. Our highly progressive tax on personal income and our 15 per cent tax on the earnings of

superannuation funds in accumulation mode are also unusual. Imputation helps ameliorate the impact of these idiosyncratic policies.

Most countries in the northern hemisphere do not tax pension fund earnings until the account-holder retires, whereupon super drawdowns are taxed in line with the regular tax rate. Thus, tax on super is essentially a progressive consumption tax. Viewing tax on super in this way, the need recedes for incessant debate about the fairness of tax concessions for superannuation. This policy also encourages voluntary contributions and self-funded retirements.

Abolishing dividend imputation without revising our current personal tax scale would result in very high effective marginal rates on dividend income, especially if the company tax rate were retained at 30 per cent. Even if the company tax rate were reduced to 20 per cent, without imputation the effective marginal rate for an individual in the 47 per cent bracket would rise to 58 per cent.

The Treasury’s Discussion Paper says that dividend imputation “makes little contribution to attracting foreign investment to Australia” and “involves a significant cost to revenue”. Thus it appears that Treasury is drawn towards abolishing dividend imputation and cutting the headline rate of company tax.

By extending the Treasury model, the study estimates that abolishing imputation could cut returns before tax to domestic investors from 7.1 per cent per annum to 6.25 per cent per annum. However, this may negatively impact retirees, as a perpetuity repriced at this yield would fall in value by 14 per cent. There could nevertheless be a rise in national income of 4.9 per cent, assuming that all the proceeds of abolishing imputation are applied to reducing the headline company tax rate.

Building on the framework of the Treasury Discussion Paper, the study further estimates that the headline tax rate for large companies could be cut from 30 per cent to 20 per cent. However, the relevant calculations rely on strong assumptions. For example, Australian companies are assumed not to revert to the low payout ratios and high gearing levels that were widespread before the introduction of imputation.

Dividend imputation may raise the cost of capital for the marginal investor, however it helps to mitigate the impact of other idiosyncrasies in our tax system. Abolishing imputation represents the capital-market equivalent of abolishing tariffs. Therefore, just as freeing up imports should be preceded by domestic liberalisation, so should abolishing dividend imputation be preceded by comprehensive tax reform.

A preferable method of incrementally promoting business investment would entail modest cuts to the headline tax rate if and when public spending commitments and the economic outlook permit.