As I mentioned a while back, I’m planning a series of posts on tax policy. Since debate about “negative gearing” has been spurred by the suggestion that Labor might restrict it, this seems like a good time to cover the topic.
I’ll give my summary upfront, then go on. The problem is not negative gearing in itself but its interaction with the concessional treatment of capital gains. There are a variety of solutions, but the best is probably to “quarantine” business losses for individuals, at least with respect to housing investments, and allow them only to be used as an offset against capital gains.
First, where does the term “negative gearing” come from? “Gearing”, in this context, refers to the ratio of debt to equity in the financing of an investment. Roughly speaking, the gearing ratio will be reflected in the ratio of interest payments to equity returns (income net of operating expenses and interest) . But what if the investment has so much debt that interest payments are higher than income net of operating expenses. Then the standard debt/equity measure of gearing is large and positive, but the second interest/equity returns measure is negative. Hence, the term.
Under the ordinary rules of taxation, interest and other expenses can be deducted from the earnings of a business enterprise before paying tax. So, there’s nothing special about the deductibility of interest in the case of negative gearing. But in the ordinary course of things, a business that makes continuous losses will close down. So, the investor has made a mistake, reducing their taxable income, and some of that loss takes the form of a reduction in tax paid. This is a substantial benefit for people on the top marginal rate of taxation, but much smaller for those on low incomes. In particular, while investment property has many characteristics that might make it attractive for retirees, the deductibility of operating losses is not one of them, since only a tiny fraction of retirees are on the top marginal rate.
The problem for revenue emerges when the investment is held for the purpose of generating a capital gain, as is the case with most investment in rental housing and many small businesses. In an ideal system of taxation of current income, capital gains would be taxed, at the full rate, as they accrued, and operating losses would automatically be set against them. Under our system, however, capital gains are taxed only when the asset in question is sold, and only half the capital gain is taxable. Not only that, but investors can time the capital gain for a year when their income, and therefore their marginal rate of taxation is low. So, there is a strong incentive to put money into investments that generate capital gains rather than those that generate income throughout their lives.
There are a variety of ways this might be fixed, such as eliminating the concessional treatment of capital gains. But the simplest (adopted, I think, in the US) is to “quarantine” business losses like this so that they can not be offset against income from other sources. When the asset is sold, the accumulated deductions can be offset against capital gains, and anything left over deducted from income.
It’s encouraging to see that Labor appears willing to tackle the problem. Overall, reining in negative gearing should improve tax and housing policy, especially given the evidence that speculative investors make bad landlords (H/T Nancy Wallace). However, a more comprehensive reform addressing issues like transfer duties land tax exemptions for homeowners would be even better.
John Quiggin 
I thought the problem with negative gearing was that losses from housing investment could be deducted from other income such as PAYE charges.
So middle class workers (middle level public servants) could reduce their overall tax if the borrowed to purchase investment properties if mortgage interest, rates, etc were not covered by the rental scheme.
In other words under Australian negative gearing the losses on investment did not have to be recovered from future earnings from that investment.
If I am correct, this is the real rort. Double income middle class families, Mum and Dad investors, will quickly see the benefit od this and immediately bid up housing prices.
Middle class groups often like to look after themselves under capitalism and have the means to do so at the cost of increasing inequality.
If we eliminate the concessional capital gains rate, wouldn’t we also have to allow investors to apply a CPI adjustment to the asset value for taxation purposes? It hardly seems fair to tax people for gains on an asset that may not have changed in value in real terms…
And if we do that, doesn’t it undo most of the concession, seeing as assets grow in real value at 2-3% with inflation at about the same level?
Negative gearing is a scam IMHO. The big winners are the big 4 banks who happily accept a fair chunk of those tax $$$ for their continual record profits. It drives up housing prices to unaffordable levels and it’s a regressive incentive as you say John – the higher the marginal tax rate, the more deductions you can claim.
I suppose if we all agreed that higher dwelling prices are good for the economy (and they could be), then yeah, there could be an argument for NG, but I personally believe the social costs outweigh any economic virtues.
Under the Howard government, “housing policy” comprised the cumulative (and perhaps contradictory) effects of: negative gearing, first home owner grants, Commonwealth Rent Assistance, and the various and continually eroding provisions of the Commonwealth State Housing Agreements. One innovative policy under Rudd/Gillard was the National Rental Affordability Scheme, which provided incentives to investors over a period of ten years in return for which the newly built property had to be rented at a discount (at least 20%) to the market rent for a similar dwelling in the area. In Queensland, tenants had to be drawn from the State Housing Register, and properties were managed by not-for-profit housing organisations. NRAS was scrapped by Abbott et al. After four rounds of incentives, just as evidence started to mount that tenants under the scheme were able to save enough to move into home ownership. Investors had to be prepared to retain the property for a decade, limiting speculation. While there were criticisms of the policy, especially as it operated in other states (and widely publicised through attack articles and commentary in the Murdoch press, in particular), it can be regarded as an example of a policy that (a) delivered a quantum of affordable housing not otherwise available (b) benefitted tenants who were struggling in the private rental market, but who were otherwise ineligible for the scarce supply of social housing (c) supported the growth of the NFP housing organisations and (d) assisted some people into home ownership. What could go wrong? …Apart from the election of a government that has yet to produce comprehensive policies for the housing sector, let alone policies for management of urban development, public transport etc.
The project: http://tenplay.com.au/channel-ten/the-project/extra/season-6/negative-gearing
@John
Part of the trick with negative gearing is to make improvements that can be expensed but contribute to capital gains. So, the typical rate of capital gain exceeds 2-3 per cent.
All of this works even better because gains aren’t taxed on accrual: this is a big concession.
At least to the extent that capital gain is not immediately assessed every financial year along with the normal earnings and deductions. It’s an obvious inconsistency to apply deductions immediately but put off including assessable income indefinitely.
@John
That used to be the capital gain formula until 1999 when the too-clever treasurer Costello decided that was unfair for dot com speculators who were making huge rates of capital gain at that time (but not for much longer of course).
Costello’s so called “reform” of capital gains tax should be reversed as Ian McAuley discusses: http://www.home.netspeed.com.au/mcau/academic/othpubs/cgtrestore.pdf
This is not the only “reform” of Costello’s that should be reversed. There is the even more obvious superannuation tax “reforms” of 2007.
@John Quiggin
Bit difficult to make improvements to shares.
“But the simplest (adopted, I think, in the US) is to ‘quarantine’ business losses like this so that they can not be offset against income from other sources.”
Under the US tax law there are two separate “quarantine” provisions applicable to individual taxpayers. First, investment interest expense is only deductible by an individual to the extent of the individual’s aggregate investment income. This provision came into effect in 1969.
Second, operating losses from a “passive” interest of an individual in a unincorporated business enterprise (including rental and mineral operations) are only deductible to the extent of aggregate income from such passive activities. The latter provision (Section 469) was enacted in 1986 and (after the drafting of complex definitional regulations) was largely successful in suppressing a serious tax shelter problem. The 1986 legislation also lowered US tax rates to a maximum of 28%, and the revenue gains from the “suppression of tax shelters served as a partial offset to the revenue loss from the lower rates.
Apart from its Australian tax implications, the term “negative gearing” sounds like a synonym for Minsky’s term “Ponzi finance”.
You could probably add the majority of family-owned farms to this sentence.
CIS Simon Cowen has unveiled a policy whereby home owners would contribute to their retirement pension.
The scheme is to access the value of their home via a reverse mortgage to supplement their pension.
This sounds like pensioners will have to subsidise the govt from their post tax savings – double taxation.
@Jim Very few if any family owned farms make a profit and/or a capital gain.
NGing copping a well deserved whacking in the press lately. Again: http://www.theage.com.au/comment/negative-gearers-are-the-opposite-of-battlers-and-they-dont-build-many-new-homes-20150427-1msy10.html
@John
The idea that the CGT system would have to compensate for inflation is a fallacy. It is a choice.
The rest of the tax system is based on nominal returns. For example, people pay tax on the full interest they receive, not just the amount above inflation – and likewise nominal interest is fully deductible for borrowers, not just the ‘real’ interest rate component.
Compensating for inflation in some parts of the tax system and not others is inherently flawed. Choosing to compensate for capital gains in particular, where there is already a substantial concession from the ability to defer and selectively realise gains and losses (as JQ refers), is doubly flawed.
(Moving to a ‘real’ tax system sounds great in theory, but would surely be a nightmare in practice.)
The better question is whether ‘investment income’ (howsoever earned) should be taxed differently (typically concessionally) compared to labour income. For example, dual income tax models.
@EconoMan
No-one claims any tax law has to be any particular way. That’s not the issue. The issue is what it should be.
That’s an attempt to turn two wrongs into a right. Doesn’t mean we should always do things wrong.
Again, you’re trying to claim that two wrongs make a right but in this case it’s even worse than with interest because the benefit from deferring capital gains tax until realization has no direct relationship at all with the cost of not adjusting for inflation.
At least we can all blame Costello for making major blunders to the capital gains tax and superannuation tax systems.
I’m sorry Chris, but I disagree. John did exactly that: his words. The routine implication from people is that if you remove the CGT discount you would have to replace it with indexation. Not as a matter of natural law obviously, but as a firmly held and potentially unconsidered opinion that no other approach is reasonable/acceptable. That is wrong.
Should the whole income tax system only tax real gains (and recognise real losses)? Sure, sounds great in theory, but incredibly complex in practice; CGT indexation and/or averaging is no exception. No tax system in the world is fully ‘real’.
Should taxation of CGT compensate for inflation any more than other parts of the tax system? This is at a minimum debatable. What is your case that it is more deserving?
Presenting this as ‘two wrongs making a right’ is also flawed. The question is what is the fairest and most efficient tax system? Given moving to a real income benchmark across the board is highly unlikely, a tax system that provides a concession (inflation adjustment) for one type of savings income and not others is unfair, and likely to be inefficient; it distorts investment (exhibit A: Australia’s property market).
The point about realisation and deferral is definitely relevant in that this already provides a substantial concession to capital gains over other savings that are taxed ‘on the way’ (e.g. interest earnings), so there is even less reason to adjust* for inflation for CGT income but not adjust for inflation for other savings income.
(Furthermore, if you plot the effective real tax rate on capital gains and incorporate reasonable assumptions of ROR and inflation, the effective tax rate generally falls over time – and the size of the deferral concession is directly linked to the difference between (nominal) return and inflation.)
* From a nominal H-S income benchmark, this adjustment is a concession. From a real income benchmark, it is neutral – but, other forms of saving are penalised. Still unfair.