In the recently released Financial Stability Review (FSR) the Reserve Bank (RBA) talked about the supply risks surrounding the inner city apartment markets however, they did point to the fact that they see more potential risk in Melbourne and Brisbane. In this blog we explore the differences in pending apartment supply across the key capital cities.
The above chart highlights the change in house and unit values across the individual capital cities since the market entered its current growth phase in June 2012. The first thing to note is that in all cities except for Hobart, house values have risen by a greater amount than units. In fact, in all of the remaining capital cities except for Sydney unit values have increased by less than half the rate of houses. This highlights that despite the shift to more unit living, particularly in inner-city areas, ultimately there appears to be a healthier relationship between supply and demand for houses compared with units and as a result, house values have increased at a faster rate than units.
Units unquestionably offer a more affordable entry point into the housing market as detailed in the above chart. Sydney’s median unit price at $675,000 is actually more expensive than the median house price in the second most expensive capital city, Melbourne ($641,500). Although units in Sydney are very expensive, you could argue that for many potential property owners they have become the only practical option for entry into the market. Looking at the differential between selling prices of houses and units, the difference is $115,000 in Brisbane, $156,400 in Melbourne and a much greater $215,000 in Sydney. In fact, Canberra is the only other capital city in which the gap between house and unit prices even comes close to the Sydney gap; it is recorded at $204,000. Urbis recently published data in their 2nd quarter Sydney Apartment Essential Report which showed that the weighted average sale price of off-the-plan units was $1,150,266 which is significantly higher than the median unit price across all stock.
The third chart looks at the proportion of annual dwelling approvals that were for units over time. Sydney has consistently approved more units than houses for construction since the 12 months to February 1993. In Melbourne, it has only been since August 2012 that the city has approved more units than houses and in Brisbane more units have been approved consistently since July 2013. Clearly the higher density living phenomena is much more ingrained in Sydney than it is in Melbourne and Brisbane. The shift to higher density in Sydney has occurred gradually over the past two decades due to the cost and scarcity of land in the city, Melbourne and Brisbane do not have similar cost or availability constraints on developable vacant land for detached housing. The other important point to remember is that an approval won’t necessarily lead to a commencement and ultimately a completion although in most cases it does.
The above chart highlights the number of units currently under construction across New South Wales, Victoria and Queensland. Unfortunately the data is not available at a capital city level however, what is occurring in each of these states is a proxy for the capital city unit markets. At the end of the June 2016 quarter, there were 55,682 units under construction across New South Wales, 46,676 under construction in Victoria and 31,070 under construction in Queensland. If you look at the long-run averages you can see that the current unit construction boom is unlike anything we’ve ever seen before. The long-run averages for units under construction are: 16,194 in New South Wales, 10,139 in Victoria and 7,429 in Queensland. While each state has seen a substantial surge in units under construction, the magnitude of the increase relative to the long-run averages is much greater in Victoria and Queensland than in New South Wales.
The above chart shows the expected increase in the proportion of units in each capital city over the next 24 months if everything that is approved and expected to be built over the next 24 months is successfully built. Brisbane is set to see the biggest uplift in total unit stock followed by Melbourne and clearly in more immature unit markets this carries a risk. Importantly though, the potential uplift in unit stock in Sydney is not immaterial at 8.5% although it is lower than both Brisbane and Melbourne.
The above table looks at the SA3 regions nationally that have the greatest number of potential new unit completions over the next 24 months. Predictably regions of Sydney, Melbourne and Brisbane dominate the list. In Melbourne and Brisbane the regions listed are generally located close to the city, within around 10 kilometres. In Sydney, the regions listed include inner-city locations as well as regions further afield, particularly along the transport spines and Western Sydney. To some extent this does disperse the risk a little more than say Brisbane where if everything is built, unit stock in Brisbane Inner will increase by 33.6%, it will rise by 33.3% in Inner-North and by 32.5% in Holland Park-Yeronga. Although, the potential risks in Sydney should not be discounted; Strathfield-Burwood-Ashfield will potentially see a 20.7% increase in unit stock, Parramatta unit stock could increase 19.2% and Auburn could see stock increase by 26.1%. While the unit settlement pipeline across the broad Sydney metro region is appears to be more manageable than the metro areas of Melbourne and Brisbane, the unit supply risks present in specific geographic areas of Sydney should be carefully considered, especially when much of the stock has reportedly been sold to offshore buyers and given that many lenders have now tightened lending policies to these types of buyers.
When you look at all these factors combined it does appear that the new unit market in Sydney is less risky than the Melbourne and Brisbane markets. Although, with the cost of new units so high and foreign buying reportedly also very high, it does not mean that Sydney is immune from these risks. Furthermore, new unit construction is spreading into new locations around the city where the market’s depth is untested and these properties are selling at prices much higher than have been seen before. While the risks may not be as great in inner city areas of Sydney as they are in Melbourne and Brisbane it will be worthwhile carefully watching the settlement performance of new units in some of the less well established unit markets throughout the city, these are more likely to become problematic than those in the inner-city areas of Sydney.
The analysis also highlights the importance of analysing supply risk at a very granular geographic level. Oversupply concerns are generally confined to specific localities. Furthermore, the risk of oversupply and non-settlement is likely to be vastly different across product types, target markets and based on the quality of the development, location of the site and the reputation of the developer.