Australia’s dividend taxation regime is a complex solution to a simple problem.

It is well known and understood that double taxation of corporate income gives rise to economic distortions.

The simple solution to that problem would be to make corporate income tax-free in the hands of shareholders.

But no. So a somewhat complicated system was introduced that taxed dividend income at the shareholder’s marginal rate with company tax effectively becoming a withholding tax.

The first version of this tax allowed franking credits to offset taxable income and excess credits were lost.

In effect this meant that taxpayers with a zero liability, and corporate income, could face a marginal tax rate at whatever the company tax rate was at the time.

The second version – introduced by the Howard-Costello government – took the policy to its logical conclusion. Taxpayers with a zero tax liability and excess credits would be refunded the full amount of pre-paid tax.

This is not a radical idea – millions of taxpayers over-pay their tax and get a refund every year.

Labor now proposes to roll back to the first version of the tax.

By Mark Ludlow and Luke Housego

A range of arguments have been proposed in support of this reform.

It’s a return to the original policy design. Cash refunds are some sort of rort. Shareholders are rich anyway. The money is better spent on public hospitals.

Only 8 per cent of taxpayers are impacted. Australians are over-invested in the Australian economy anyway. This latter argument is particularly strange – an Australian government would be discouraging local investment.

The questions that should be asked are “Is the revenue actually there?”, “Are there behavioural responses that will undermine the policy?”, “Where is the burden of the tax going to fall?”.

In an attempt to answer these questions I had a look at the ATO individual taxation statistics for 2015-16 (the latest available year) and specifically at their individual sample file.

This data file provides detailed tax information for a representative sample of the taxpaying population.

Revenue expectations

According to the ATO data some 21 per cent of the Australian population report receiving franked dividends in that year.

I then used the tax tables to calculate an imputed tax payable for each taxpayer and compared that amount to the amount of franking credits.

Some 6.1 per cent of taxpayers appear to earn excess franking credits – less than the 8 per cent Labor claims. That could be good or bad for Labor’s expectations as to revenue.

So the next question is, who are these 6.1 per cent?

Women make up 56 per cent of taxpayers earning excess franking credits but only 48 per cent of the sample data.

Of those women, 68 per cent are over 60. Labor wants to tax your Nana. But maybe Nana is rich. Maybe Nana and Pops have organised their affairs so that the share portfolio is in Nana’s name so as to minimise their joint tax liability.

So I went and had a look at the total and taxable incomes for all women and then those women with excess franking credits.

Taxable and total income figures for those women with excess credits tended to be lower than all women on average, and for each age group.

To be certain that this wasn’t due to tax planning I then added the franking credits back to total income and recalculated all the average and found the same result.

These women are not “rich” and, on the assumption that not reporting spousal details indicates single status, some 47 per cent of these women are either single or widowed.

So a whole lot of (widowed) elderly women are about to face a marginal 30 per cent tax rate under a new Labor government, and will lose a portion of their current income.

The numbers go down once I account for the pension guarantee that Labor has proposed – but not by much. We are being invited to believe that this tax grab on the elderly won’t have any flow-on effects to their children and loved ones.

What could these women do to “avoid” the tax on their livelihoods?

I expect that many will sell down their share portfolios to qualify for the pension (or for more of the pension) and use the proceeds to renovate their homes.

I doubt that this is the policy intent, but it does suggest Labor will raise a lot less revenue than it imagines, and that the social costs of this policy are higher than they anticipate.

Sinclair Davidson is a professor of economics at RMIT University, an adjunct fellow at the Institute of Public Affairs and an academic fellow at the Australian Taxpayers’ Alliance.