Back to our regularly scheduled programming…
Now that we have data in a scriptable form, we can look at how the market has behaved when various market conditions are in place. Two of the most commonly cited metrics for determination of whether a market is tilted towards sellers or buyers are the sales/list ratio and months of inventory. The theory being that if the ratio of sales to new listings is high, or if the months of inventory is low, then demand is outstripping supply and prices should be rising. For now we’ll use the months of inventory, my theory being that it is a more accurate reflection of current demand than the Sales/List which might be more of a leading indicator (Sales/List could be very high but if there’s enough inventory that doesn’t mean there is a pressure on prices). However this deserves another look in the future.
So how do prices respond relative to months of inventory?
Unsurprisingly, there is some kind of relationship there, with lower months of inventory leading to higher price increases. Note that I’m using the 12 month trailing MOI to remove seasonality, and the price increase in the following quarter extrapolated to a year to calculate rate of annual appreciation. To put this into more of a usable form, we can bin the months of inventory and chart the percentiles (solid blue is 25%-75% percentile, whiskers are max/min).
So a couple takeaways:
- Victoria’s history shows lower months of inventory lead to higher price increases and vice versa.
- There is still a lot of variability in price changes for a certain MOI value. For example, our trailing 12 month MOI is currently 4.13. In the 4 bin, we have seen price appreciation at a rate of anywhere from 5% to 15% annually.
- There is a definite “stickiness” about prices at higher months of inventory. It seems the linear relationship breaks down above about 7 MOI.
The price predictions on the stats page are based on a similar binned chart as above. Using the current 12 month trailing MOI, we take the median of all price movements in the past with the same MOI and use that to predict the annual movement in the future. Now of course this is subject to a lot of limitations, the most obvious being that markets are constantly changing, so the chance of market conditions staying the same are about zero. It ain’t perfect, but it’s a start at an evidence based and unbiased projection of prices.
Some areas of future investigation
- Examining the relationship between sales/list and price. With a similar method, how closely would they predict price movements?
- Using another method that doesn’t rely on 12 month trailing averages to remove seasonality. By using trailing averages we are likely being a bit too heavy handed and gathering a lot of noise with the data.
So does anyone have a theory about why prices in Victoria so sticky? Even at hugely high months of inventory that would lead to price declines in other markets, our prices barely drift lower. Do we need higher levels of unemployment to see price drops?