Read the paper at www.cis.org.au

Executive Summary.

This paper is an expanded version of a submission to the Senate Select Committee on the Operation of the Capital Gains Tax Discount. The author gave evidence to the Committee at a hearing on 25 February, 2026.

Although there is much public discussion of the capital gains tax discount, there is no proposal from government on the table for us to respond to — only rumours and speculation — so our comments are broad-ranging and not confined to housing.

As well as the submission, there have been three relevant research publications on CGT issued by the CIS in 2009, 2015 and 2019. Perusal of those publications will show that we do not think much of proposals to reduce the CGT discount.

If three publications looks like an obsession, we have had a lot to say on the issue because calls for the discount to be cut or eliminated have been a persistent theme of tax policy debate ever since the defeat of the Howard government, which put the 50% discount in place in 1999. Along with superannuation concessions and negative gearing, the discount has been a favourite whipping boy.

Cutting the discount is variously seen as a key plank of tax reform, a revenue-raising measure, the key to lowering house prices, and a solution to intergenerational and vertical inequality. Our submission argues that it is none of those things, or at least not in significant measure, and that the 50% discount is justified.

In brief, we make the following points:

The principle of taxing nominal capital gains at lower rates than ordinary income is unexceptional and was recognised in Australia’s first model of CGT in 1985. The 50% discount in 1999 replaced what was essentially a different form of discount in the 1985 model based on indexing the cost base of assets to CPI inflation combined with an averaging scheme that limited the effect of lumpy capital gains pushing taxpayers into higher tax brackets.

The pre-1999 arrangements produced a variable discount, but for average rates of return on assets and inflation rates and various asset holding periods it can be demonstrated that the effective discount often fell in the 30–45% range — and that is leaving aside the additional benefit of averaging. The point is that the 50% discount is not much more generous than the average result of the policy it replaced.

To those who say the 50% discount over-compensates for inflation, in some cases it does and in some it compensates or even under-compensates, but the key point is that it was never intended solely to compensate for inflation. It was meant to be a general incentive for saving and investment, which is needed now more than ever in view of stagnant productivity.

On housing, several researchers have estimated that cutting the CGT discount would reduce house prices by a few per cent while increasing rents by a similar amount. These effects are tiny relative to other influences on prices and rents. CGT affects much more in the investment world than housing, so housing considerations should not drive CGT policy.

The claimed revenue costs of the discount vastly overstate the revenue that could be gained from any reasonable change. The claimed distribution of that revenue cost across income deciles is meaningless. Cutting the CGT discount would barely move the dial on income and wealth distribution.

Changing the CGT discount on its own is not tax reform, but it could have a place in broad tax reform that substantially reduces marginal rates and reduces the large disparities in the tax treatment of different forms of saving. However, nobody in government is talking about that.

The key conclusion is that there is a very strong case for some form of tax concession for capital gains relative to full marginal rates. This concession should go beyond simply allowing for inflation. While various structures are possible to satisfy this condition, the current 50% discount (and one-third discount for superannuation funds) has the advantage of being simple and well understood. It has been the basis for investment decisions over the past 26 years and is therefore entrenched in the accumulated stock of investments. There is no strong case for changing it.

Read the paper at www.cis.org.au