The Devil Is In The Data Detail

The modern world of technology serves up a banquet of information. Few things whet the appetite of the avid property investor more than a good serve of data. But, before you hook in to your feast, it is important to understand exactly what’s on your plate. Unless you understand the ingredients you may choose a dish that has a bitter taste.

Much of the data published in the mainstream property media is quite generic. Common examples include recent changes in median property values, sales volumes, vacancy rates and auction clearance rates. While this type of data can be interesting, it doesn’t greatly excite my taste buds if I’m contemplating whether or not to invest in a specific market. The decision that I make to invest today is one which I’ll be stuck with several years so it’s not greatly relevant to me how a particular market performed last year or last month.

I prefer to process information with consideration for how each it may influence one of the three pistons in the property market engine: Sentiment, Demand, Supply.

Sentiment, or the ‘mood’ of a community, directly affects consumer spending habits and decision-making. Unfortunately, this is very difficult to measure, especially at an individual city level. Australia’s Big-4 banks each produce separate reports which contain some useful data on the Sentiment side of the equation at a state-by-state level. Westpac has a consumer confidence index, CBA has created a series of metrics in their CommSec State-Of-The-State’s Report, NAB has some good data in its Business Confidence Report, and ANZ measures job advertisement data.

Given that property can be such an emotional decision, the number of properties listed for sale, volumes of sales, and how long they take to sell are reflective of a community’s general appetite. Trends of each of these sets of data can be a precursor of price trends in the near-term however, they bare little relevance to medium term fundamentals of a market.

On the Supply side of the equation, vacancy rates are the most commonly referred to set of data. Provided the data sample size isn’t too big (aka, a vacancy rate for an entire capital city) or too small (an individual suburb) vacancy rates are useful as a landlord when trying to determine the level of rental pressure when a lease is up for review. Generally speaking, 3 per cent is ‘normal’ while pressure to increase rents starts to mount once vacancy rates dip below 2 per cent. Vacancy rates will go up and down like the cost of a kilo of bananas over the years so they are not an essential ingredient when I’m deciding where to invest.

The most important data in respect to housing supply is building approval data. From 2013 onwards, residential construction in this country has never been higher and reports of over supplied property markets across the country are becoming increasingly common. The leading indicator for future supply is building approvals so Propertyology monitors the trends of each of Australia’s 550 city councils. The gestation period from building approval to completion on bigger housing projects can be a couple of years. To get the right balance of sweet and sour, I find it very useful to compare approval volumes of the last couple of years to annual averages of the previous ten years.

There’s several more important considerations relating to housing demand than population growth – believe me, I’ve studied the historical evidence! More important than population growth are availability of work for the population, what the population can afford, and how much new housing the population builds.

Amongst the hundreds of spreadsheets in Propertyology’s data library is interstate migration statistics. It’s a big decision for someone to uproot their life and relocate; the two most common reasons for doing so are employment opportunities and / or housing affordability. Sustained increases and decreases in interstate migration is often reflected in capital city property market performances. One of the better case studies for this was Brisbane in 2002 to 2004 when an influx of (mostly) former Sydney residents drove Brisbane property prices sky high.

Housing finance data (albeit only available on a state-by-state basis) can provide some important insights about property markets. Upward and downward swings in loan transaction volumes is a direct indication of buyer mood; the level of competition is often the leading indicator for price fluctuations. Deeper analysis of finance approval data can reveal trends in investor loan activity. For example, approximately 65 per cent of all residential loans approved in New South Wales (aka ‘Sydney’) in 2015 were for investment purposes and logic would suggest that this significant increase in size of the rental pool will release pressure on rents.

As a self-confessed number nerd, I can’t get enough data about the economy (federal, state and local). I like to categorise the economic profile of each of Australia’s 550 local government jurisdictions with data that includes the percentage of the workforce employed by each industry sector, the average household income, and average household age. Following the monthly trends in job numbers (which I believe are more important than unemployment rates) at a local government level can be a good indication as to future demand for housing. Remember, wherever there is demand for more jobs in the future there will be demand for more shelter.

Data company publications are very capital city centric; hardly a week goes by without one of them serving up the same boring dish. Of the 550 local government jurisdiction in Australia, 139 of these (only 25 per cent) are in capital cities whereas some of the most exciting opportunities for property investors are found amongst the 411 regional markets (there’s some very tasty dishes amongst the 75 per cent of regional Australian markets).

When analysing data, it’s important to remember that past performance is not a guarantee of future performance (in fact, quite the opposite may happen before too long). It’s important to understand that certain types of data does little more than satisfy one’s curiosity about the past (for example, a change in median property values). Other data might relate directly to the present (such as auction clearance rates and vacancy rates). For me, the juiciest data relates to leading indicators about what lies ahead (such as building approval volumes and employment-related data).

It’s important to have context to the data that you’re analysing. For example, comparing data of a specific location to a national and state average. Often there might be more value in understanding the trend rather than the actual number itself. Perhaps the best example of this is Hobart’s property market which, until recently, had been totally removed from most investor’s considerations due to Tasmania’s very high unemployment rate. Those of us who dug in to the detail of the data quickly discovered that Hobart’s unemployment rate was considerably better than the state average and that job growth trends were improving. After acquiring further understanding of Hobart’s economic drivers, I purchased property in Hobart in mid-2014 at a time when the market was completely flat. Fast forward two years and there’s a range of fresh data which now paints Hobart as one the strongest markets in Australia.

Before looking at how different sorts of data can be useful for analysing property markets, it’s important to understand that data doesn’t tell us everything that we need to know. You’ll have to travel a lot further than to your pantry to find precious ingredients such as the outlook for an industry, a key announcement by a major employer, plans for a significant piece of infrastructure, and zoning changes.

Remember, it’s the sum of all of the ingredients in the dish, not one specific ingredient, which will determine how you rate your dining experience.