My plan to cut tax is feasible

What are we trying to achieve with our tax system? We should be endeavouring to raise sufficient money for the federal Government's needs in the simplest, most efficient and equitable fashion that we can design.

We know that taxes impose a deadweight cost on the economy over and above the expense of the tax. So we should not impose taxes at any level that is higher than it absolutely needs to be.

Tax reform is not just about tax cuts; at any level. Tax reform involves creating a simpler and more efficient system. Lowering the rates goes hand in hand with broadening the base. This means eliminating many of the concessions and loopholes that make tax compliance more expensive and wasteful of time and effort.

But most of the discussion this week has been about the top rate: the 47 per cent tax bracket. Only 3 per cent of taxpayers will be in that bracket next year. So, the political argument goes, we may as well slug them. Who wants to be seen letting the millionaires off lightly?

But this is just political rhetoric. Anyone with substantial business income or interests is able right now to structure their affairs so that most of their income is earned through companies, which pay tax at 30 per cent. Our tax system has been designed to ensure that, by and large, the only people who pay the top rate of tax are, as the Melbourne Institute recently confirmed, PAYE taxpayers or professionals.

That, of course, is why the top rate raises so little money. Indeed, on our calculations, to remove the 47 per cent bracket completely would at most cost in 2006-07 less than $450 million; less than one-quarter of 1 per cent of the federal budget. And that is before we take into account any improvement in investment, participation, compliance or productivity. The projected costs of reform in our modelling are for these and other reasons very conservative and likely to overestimate the costs of tax reform. In the light of those factors, we could be very confident that removing the top rate would probably pay for itself.

In other words, there is simply no financial or economic justification for maintaining the 47 per cent rate. Its only justification is political: "soak the rich". But such a rationale is completely disingenuous. After all, anyone who is not a PAYE taxpayer can easily and legally structure their affairs so they don't have to pay it.

I have not, however, advocated simply removing the top rate of tax. I also believe that our tax system should be improved by an across-the-board reduction in tax. This would mean all taxpayers would benefit, in terms of less tax and a simpler tax system that is easier to understand and comply with.

The paper Jeromey Temple and I published last week was designed to inform and stimulate debate rather than put forward a specific policy proposal. We simulated, therefore, the fiscal and distributional consequences of a very large number of possible changes to rates and thresholds so that the Australian public would have a good basis of information with which to participate in a debate about taxation.

However, there are a number of the possible models that are obviously more attractive than others, either because their cost is low and-or they deliver a tax cut to a broader range of taxpayers. We particularly focused on those reforms that cost less than $10 billion; a sum that can readily be found from eliminating concessions and improved productivity, compliance and participation.

It is also a sum not much more than the size of the forecast surplus. We do not assume tapping the surplus to pay for tax reform, although there is no reason it should not be available, at least in part, to fund a more efficient tax system.

Without wanting to advocate, at least at this stage, a particular set of reforms, we can focus on a couple of the more feasible reforms as a basis for discussion.

A system with three brackets: 15 per cent, 29 per cent and 40 per cent would cost, at most, about $3.8 billion in 2006-07. It would deliver almost all taxpayers a cut. The top 20 per cent of income earners would get a 4.5 per cent cut, the next 20 per cent a 3 per cent cut. Even the bottom 60 per cent (who on our figures will pay about 15 per cent of the total tax take in 2006-07) would get a tax cut. Because most taxpayers are in the 30 per cent bracket, the 1 per cent cut in that bracket would cost about the same as the 7 per cent cut in the top brackets.

As we discuss in our paper, means-testing of benefits has the consequence that welfare is withdrawn as income is earned. The combination of a withdrawal of benefits and the imposition of income tax can result in high effective marginal tax rates for low-income earners moving from welfare to work. There is no silver bullet for this problem; it is inherent in any means-tested welfare system. But some relief can be given by lowering tax rates in the bottom bracket (as was done in the previous budget with the reduction in the 17 per cent rate to 15 per cent) or by lifting the tax-free threshold.

Lifting the tax-free threshold is very expensive because it benefits every taxpayer, not just those on low incomes. But one reform package that delivers broad tax cuts would involve raising the tax-free threshold to $10,000, and then dropping the top rate to 40 per cent. That would cost (at most) $7.5 billion. The top 20 per cent would get a 5 per cent tax cut, as would the next 20 per cent. The last 60 per cent would get a 26 per cent tax cut. And of course EMTRs would be reduced.

If the top rate reduction was increased to bring it down to 35 per cent, the cost would increase to $11 billion and the top 20 per cent would get an 11 per cent tax cut, the next 20 per cent a 5 per cent tax cut and the last 60 per cent would get 26 per cent tax cut.

But because of the big impact a change to thresholds has on those on lower incomes, the share of total personal income tax paid by the top 40 per cent of taxpayers would increase.