The Australia Institute is in the media this week with a proposal to tax capital gains on high-end owner-occupied housing. They’ve backed the call with modelling that shows the current exemption of owner-occupied housing (all owner-occupied housing, not just the high-end stuff) benefits high-income households the most: of the $46 billion in tax revenue foregone this year because of the exemption, more than half of it is kept by the top 20 per cent of households by income, with more than one third kept by the top 10 per cent. Under the Australia Institute’s proposal, rolling back the exemption from properties sold at more than $2 million would affect just one per cent of properties, but raise almost $12 billion in tax over the coming four years, mostly from the top 20 per cent of earners.

The Institute presents its case primarily in terms of ‘budget repair’ and encouragement of productive investment. For some reason it buries what is the best argument for rolling back the CGT exemption: it would help make housing more affordable (this gets just a single mention in their paper). Housing is overpriced because it is undertaxed.

The most important thing about tax exemptions is not their estimated dollar value, but their influence on people’s behaviour (the dollar value can give you a sense of how powerfully it is influencing behaviour, but if the behaviour were different the value would be different too). And what the CGT exemption for owner-occupied housing does is encourage households with money to spare to throw that money at their own housing, in preference to other classes of asset (for example, a share in business), where any capital gains will be taxed (at half the money-thrower’s marginal tax rate).

Owner-occupied housing gets preferential tax treatment over other assets in another, often forgotten way. An asset such as a share in a business, for example, may generate, from the valuable goods and services it produces and sells, an income for its owner, which will be taxed at the owner’s marginal tax rate. Owner-occupied housing produces something valuable too – shelter – but this is consumed directly by the owner, without paying any tax on it. The dollar value of the exemption for this ‘imputed rent’ is not regularly reported by the Australian Treasury (as noted by the Australia Institute in another paper); in 2013, the Grattan Institute estimated it at about $12 billion.

So that’s more encouragement to people to throw money at owner-occupied housing. And where owner-occupiers are throwing money at houses, there will be people buying houses not to live in themselves, but to potentially catch some of that money that gets thrown around.

These so-called property investors face a different range of tax settings: unlike owner-occupiers, they’re liable to pay CGT at the usual discounted rate, and they’re liable to pay income tax on their net rental income… except, two-thirds of them operate at a loss after deducting interest and other costs (ie negative gearing), because also unlike owner-occupiers, they can deduct these costs from their other incomes. These settings don’t treat investment properties differently from other asset classes, but they do preference high income earners over low, high gearing over low, and capital gains over income. And when you set up high income households to gear into the pursuit of capital gains, and set up a particular asset – houses, that can trade in both investor and owner-occupier markets – with tax exemptions that encourage owner-occupiers to throw money at it, a whole lot of money gets thrown from all corners… even to the extent, as we’ve seen recently, that the money thrown by the so-called investors is more than the money thrown by owner-occupiers.

Reforming the tax-preferenced status of owner-occupied housing is one of the keys to making housing more affordable – particularly for those would-be owner occupiers who have been priced out in the great housing money-throwing fight. A couple of complications should be noted, however, arising particularly from the way in which the owner-occupied house has become a mix of a home for a household and a vehicle for speculation. In so many cases, when a house is sold all the proceeds go into the purchase of another house. If CGT was payable, we should expect an effect like that of stamp duty – a fine for moving house that discourages useful and desirable moves. There is also the question of how interest payments and other costs should be treated: if housing really was to be taxed like other asset, interest would be deductable too.

Those concerns are less of a problem considering the limits of the reform proposed by the Australia Institute. By limiting the rollback to $2 million+ properties, the proposal targets those houses where the owner’s speculative vehicle/trophy purposes are larger than their home-making purposes, and counters the tendency for high-gearing investors to go into ‘premium’ properties, rather than the lower value, lower rent stock that has gone missing from the market. The proposal deserves consideration by the Government as a measure for greater housing affordability.