Are too many houses empty?

Earthsharing Australia - Suburbs with highest proportion of properties withheld from rental market (last col)

Earthsharing Australia highlighted this week what could be a major housing supply issue in Australia’s major cities: the number of houses sitting vacant at any one time. Properties will inevitably be vacant from time to time – that’s necessary for an efficient market – but the issue is whether there are structural factors that mean too many are unoccupied for too long.

Earthsharing has attempted to quantify the number of unoccupied dwellings in Melbourne. It claims speculators are the reason why five per cent of the city’s housing stock – or 46,220 dwellings – sit empty and unrented at any time. It says the REIV’s Rental Vacancy rate is commonly referred to in media coverage as the ‘housing vacancy’ rate, but Earthsharing’s own:

Estimated Speculative Vacancy Rate (of 4.9%) is more than twice the REIV’s Rental Vacancy rate for the same period of 1.7%…….Recent increases in house prices have been driven by speculation, not a housing shortage. Property buyers are restricting the supply of housing by holding their properties off the rental market.

The Estimated Vacancy Rate for some suburbs was much higher – in Docklands it was 23% and in East Melbourne, 19% (referred to in the exhibit above as Estimated Genuine Vacancy rate).

Earthsharing’s findings are contained in a report released on the weekend, Speculative Vacancies in Melbourne 2010, which measured the number of houses (excluding the area covered by South East Water) that consumed less than 50 litres of water per day, on average, over a period of six months in 2010. The presumption is dwellings using less than this amount must be unoccupied, given that average daily consumption during the period of the study was 140-160 litres per day. The further assumption is these dwellings have been withheld from the rental market for speculative reasons.

Earthsharing’s Speculative Vacancy rate could be conservative. Unoccupied dwellings with an automatic sprinkler system or a serious leak might consume more than 50 litres per day and hence be under-counted. On the other hand, its methodology could over-count the number of unoccupied dwellings. There’s some evidence consistent with the latter view – of the 46,200 properties identified by Earthsharing as “withheld from the market”, only 15,237 consumed zero litres of water over the six month period, and hence could be regarded as unambiguously vacant.

In fact there are many reasons why a property might be occupied but nevertheless average less than 50 litres per day over six months. Apart from rental dwellings between leases, some properties are unoccupied because they’re being sold by one owner-occupier to another. There are city properties owned by country people who use them regularly but relatively briefly e.g. a weekend every fortnight. Then there are single person households who travel frequently or for extended periods, as well as “couples” where each party has their own home but they favour one.

It might be possible to refer to these sorts of properties as “under-occupied” in the same sense that empty nesters rattling around in four bedroom houses is “inefficient”, but it would be a big stretch to label them with a pejorative like speculative. These aren’t properties that are being withheld from the rental market. In short, Earthsharing’s methodology doesn’t seem very robust.

But having said that, I suspect there are far too many non-rental properties that sit unoccupied for unnecessarily long periods. Let me emphasise that I don’t have any objective data to support this contention, but if it’s right, it would add to pressure in the rental market. Consider that within 500 metres of my place (I live 8 km from the CBD) there are four properties I’m aware of that have sat vacant in recent years for twelve months or more.

In the first case a property with redevelopment potential was sold for almost $3 million to a developer and until recently subsequently sat vacant for two years. The new owner presumably spent time examining multi-unit options until seemingly giving up and recently putting the property back on the market.

The second case involves a single parent family renting a property owned by a government infrastructure agency. They were given notice to vacate so the property could be sold. The property then sat vacant for about twelve months before being refurbished and re-let (presumably by a new owner). What I don’t know is if any time was spent investigating redevelopment options.

The third case is a newly constructed ‘spec’ detached house that was passed in at auction then sat vacant for at least twelve months while the developer sought a buyer (actually, I’ve never seen any signs of life so I’m not sure it’s ever been occupied!).

The fourth case is a newly constructed small-lot townhouse that’s 99% complete but has remained unfinished and hence unoccupied for more than two years. I don’t know what’s happened here – it could be a dispute between the builder and the developer – but it’s a property that’s not being put to productive use. Only a small amount of work would be required to enable a household that’s currently renting to live in this house.

Whatever advantage there is in leaving these properties unsold and/or unrented for such long periods – and, presumably, thereby foregoing some tax advantages – I don’t think the term speculative is warranted. Established properties where the significant value of infrastructure and services is already capitalised into the purchase price seem a very different kettle of fish to the more familiar notion of speculators on the urban fringe buying broadacres cheap and letting them sit fallow in expectation of future “windfall gains” from urban development.

Perhaps developers under-estimate the time involved in investigating development options and getting approvals. Maybe they think their obligations to renters are too risky or onerous to even bother leasing properties on a twelve month term. Possibly the cost of capital is so low for some non-resident owners they can’t be bothered with the hassle of renting their properties or alternatively they don’t benefit under the Australian tax system. Contractual disputes that leave assets unused for long periods are hopefully not too common but we should be aware there can be social costs as well as private ones. Whatever the reason, it seems there could be a significant proportion of housing that is simply sitting empty.

The Census next month will tell us what proportion of the housing stock is unoccupied on the night of 9 August 2011, but it won’t tell us what proportion is in the long-term unoccupied category. If we knew, we might be surprised how large it is.