Sydney lurches to housing affordability disaster

Most of the Big Projects coalition and worker-trader class subscribe to a supply-solution in principle. But there are differences on what this means in practice. An explicit apportionment of new housing between brownfield-infill sites and greenfield development was dropped from the NSW Government’s A Plan For Growing Sydney. An outcome is achieved by focusing on 14 Priority Growth Areas and Precincts, only 5 of which contain substantial greenfield potential.

Growth Areas inside established built-up localities are Rhodes East, St Leonards and Crows Nest, Greater Parramatta to Olympic Peninsula, Sydney Metro Northwest, Sydenham to Bankstown Corridor, Western Sydney Employment Area, Epping and Macquarie Park, Arncliffe and Banksia, and Ingleside Precinct. The outer, peripheral areas with most capacity for new land release or greenfield development are North West Growth Area, South West Growth Area, Greater Macarthur Growth Area, Western Sydney Growth Area and Wilton New Town, which lie within seven Local Government Areas (LGAs); Blacktown, The Hills, Camden, Campbelltown, Liverpool, Penrith and Wollondilly Shire.  

Under A Plan for Growing Sydney, the authorities are planning for an additional 1.6 million people and 664,000 dwellings across the Sydney metropolitan region by 2031. According to NSW Department of Planning “state and local government area household and implied dwelling projections” to 2031, Blacktown LGA will have 48,300 new dwellings, The Hills LGA 28,650, Camden LGA 38,250, Campbelltown LGA 19,450, Liverpool LGA 32,400, Penrith LGA 20,900 and Wilton New Town, in Wollondilly Shire, 16,000 dwellings. In other words, these fringe priority areas are to accommodate an additional 203,950 dwellings, or around 30 per cent of the extra 664,000 dwellings across metropolitan Sydney.

Of course, not all construction in the 7 peripheral LGAs will be on new land, so the share of total dwellings on greenfield sites will be even lower. The Urban Development Institute of Australia (UDIA) estimates that Sydney greenfield lot production is running at 11,600 a year and will reach 12,355 a year in 2017/18. If achieved, that translates to 185,325 or 27 per cent of the 2031 metropolitan dwelling forecast (equating a fringe lot to a single dwelling). 

This month, the Department of Planning released accelerated forecasts totaling 184,300 new houses and apartments across the 33 metropolitan LGAs by 2021. Of these, 8,350 are assigned to The Hills LGA, 13,600 to Blacktown LGA, 11,800 to Camden LGA, 6,700 to Campbelltown LGA, 8,050 to Liverpool LGA, 6,600 to Penrith LGA and 1,450 to Wollondilly Shire. Together, these represent 30 per cent of the metropolitan total. Large increases are channeled into established areas, including 21,450 in Parramatta LGA, 18,250 in Sydney LGA (covering the CBD and surrounds), 12,200 in Canterbury-Bankstown LGA and 10,000 in Bayside LGA. NSW Planning Minister Rob Stokes boasted “we are getting the balance better … getting over the greenfield issue was the biggest thing that needed to be done”. The targets were to be fleshed out in Draft District Plans administered by a new planning politburo, the Greater Sydney Commission (GSC).  

Within days, however, the GSC announced its own strategy and targets. The total housing target is distributed to 6 Districts across the city, Central, North, West Central, West, South West and South, rather than Priority Growth Areas. Based on a metropolitan total target of 725,000 dwellings for 2 million more people, each Draft District Plan nominates a 20 year target to 2036. The South West District contains 4 of the peripheral LGAs with most potential for greenfield construction, Camden, Campbelltown, Liverpool and Wollondilly. Its 20 year housing target is 143,000 dwellings, or 19 per cent of the metropolitan total. Of the other 3 LGAs with most greenfield potential, Penrith accounts for just part of West District’s target of 41,500 dwellings or 5 per cent of the metropolitan total, while Blacktown and The Hills are in West Central District, which is dominated by Parramatta LGA with minimal new land release capacity.

Higher the density, higher the prices

Suppression of greenfield development reflects a view that location and density don’t condition the benefits of supply. Yet this is contrary to a body of economic analysis on the land value impacts of urban containment. Citing LSE economist Paul Cheshire, commentator Phil Hayward gives a cogent account of this in “The Myth of Affordable Intensification”.

Hayward explains that the more density allowed, the higher the average housing unit price becomes. Cheshire put this down to a bidding-war at the margins of each income-level cohort of society for slightly more space. The less average space available per household, the more intense is the bidding-war effect. Site development potential in an urban land market with a regulatory limit on land supply, writes Hayward, seems to capitalise into site values. When the market allows people to consume as much space as they want, the bidding-war effect is absent.

Urban land economists like Cheshire and Alan Evans at Reading University consider housing a complex good … consisting of many attributes bundled into one composite good. The land base is a particularly important attribute. With rising population and incomes, restrictions on the quantity of land at the periphery ratchet up values across the whole urban region. The evidence that fixed urban growth boundaries put upward pressure on land and thus house prices is clear. While no formal boundary is proposed for Sydney, delimited Priority Growth Areas and GSC Districts have the same effect, operating as land value traps. Between 2009 and 2014, the Sydney median greenfield lot price ballooned from $269,000 to $339,750, reports the UDIA, even though lots released per annum rose from 2503 to 8597.

To subdue prices, Cheshire argues in a 2009 paper, it isn’t enough to rezone and release enough residential land to meet anticipated demand:

If we are to provide stable prices … what we need to predict is the effective demand for housing and garden space given that it is the quantity of land that the system allocates. Then we have to allocate not just the quantity of land predicted as being compatible with price stability but more. Not all the land allocated as available for development will actually be developed. One rule of thumb suggested is that this implies allocating 40 per cent more land than the estimated demand indicates is needed.