The other day I posted about the Henry Review’s central recommendation for the personal income tax system. At the Liberal Party’s ‘policy launch’ Tony Abbott dropped some broad hints about tax, promising to “review the Henry Review”, which apparently is what passes for policy nowadays. Peter Martin thinks that Tony Abbott “squibbed it”, that he planned to announce a tax policy over the weekend but decided to remain a small target instead. He’s probably right.

Peter Martin also kindly linked to my post, but seemed to suggest that a regressive tax cut with big dollar gains at the top end was no cause for concern if those gains were no larger as a proportion of income than the gains at the bottom end.

I disagree.

I am an unashamed supporter of a progressive income tax system. I support progressivity for the usual social-democratic equity reasons; I believe that the relatively well-off have a moral obligation to contribute a greater proportion of their income in tax than the less well-off.

But I also support a progressive tax system for reasons of efficiency, as outlined by Richard Green at Club Troppo. Green outlines a lot of reasons to support progressivity on efficiency grounds, including the important fact that taxes on higher income earners distort their behaviour less than an equivalent tax on lower income earners, due to diminishing marginal utility. Put simply, an additional dollar of income is valued more highly by low income earners than by high income earners.

I’d like to focus here on one particular argument in favour of maintaining the progressivity of the tax system, namely the role of marginal tax rates in encouraging, or discouraging, the supply of labour. In particular, I’d like to ask: who would have an incentive to increase, or decrease, their desired hours of work (which we’ll define as ‘labour supply’) if we moved from our current personal income tax structure to the one proposed by the Henry Review panel?

The answer: people at the very bottom and the very top would, all other things equal, be expected to increase their expected hours of work, while significant numbers of people in the lower-middle would be expected to cut back.

The green shaded areas in the graph above show my estimate of where the current marginal tax rate (MTR) is greater than the rate under the Henry structure. Over these spans of income, taxpayers would be likely to increase their labour supply if the Government adopted the Henry Review’s recommendation. The red shaded areas show the opposite, areas in which the Henry MTR exceeds the current MTR.

It’s clear that there are gains at the bottom end. The effect of the Henry recommendation on the low income earners in the $16 000-$25 000 per annum range is very important, and these are the very people we should be most concerned about. They’re likely to be the most responsive to a reduction in their effective MTR; in economists’ jargon, their own-price elasticity of supply is likely to be greatest.

However, there are a lot of people in the $25 000 to $37 000 range who would face a higher MTR under the Henry structure than under the current system. This includes a lot of second earners, a lot of part time workers, a lot of people we should be concerned about when designing an optimal tax system.

Notwithstanding the significant gains at the very bottom, I fail to see the logic of advocating a system in which a substantial span of low-middle income earners would face a higher marginal tax rate and thus be induced to reduce their labour supply. Note that it’s not just a question of marginal rates either; I showed in my last post that many people over a span of roughly average income plus or minus $20 000 would face a higher average rate, too, and therefore pay more tax overall.

With any income tax reform, there is a finite amount of foregone revenue that can be allocated among taxpayers. My belief is that that revenue should be allocated where it will have the most impact in encouraging labour supply, boosting utility, generating revenue with minimal distortion, and redistributing income. I fail to see how these objectives are met by a system that imposes losses on the lower-middle and middle while distributing substantial gains to the top.

(Incidentally, the dollar AND percentage gains at the top end are much bigger when you include the Medicare Levy Surcharge, which I haven’t done here. My analysis includes the 1.5% levy, but not the 1% surcharge).