Each month Tom Ruff of The Information Market gives his stellar commentary on the housing market. Tom is armed with Pending data of which others do not have access. His insights are below. Read the full issue of STAT for the accompanying graphs.
The charts in STAT are mostly self-explanatory but sometimes they need context when market conditions change or even when the calendar falls funny. July 2016 was one of those months. On the surface, MLS sales numbers were disappointing with volume down 3.6% year-over-year and down 13.9% month-over-month. However, the decline in sales is more a factor of the calendar as opposed to declining demand this month. If you compare the sales volumes of July 2015 to July 2016 in the context of business days, you have your culprit! There were 359.7 sales per day in 2015 and 381.5 sales per day in 2016.
I have already seen early reports talking about how market conditions contributed to slowing July numbers. When viewing the closing numbers in July I can almost guarantee you, the missing sales in July simply got pushed into August closings. When August numbers are reported, they’ll be talking about stronger numbers but for both months it was simply the calendar.
Over the past few weeks our team has been working on various Automatic Valuation Models. An AVM is mathematical modeling based on market forces to make predictions. I would like to share with you some of our early observations derived from our modeling efforts.
Zestimates can sometimes cause uproar with agents. We also create models but make no mistake; no model can outperform the price opinion of an ARMLS Subscriber on an individual property basis, period. If you want to know what your property is worth, ask an ARMLS Subscriber. Our analysis is based on listing prices and at the moment properties are selling at 97.5% of the list price on average. Our model can’t take into consideration the condition, curb appeal or other factors at play but we need to work at making predictions to better help our Subscribers.
The Full Cash Value (FCV) as determined by the County Assessor is based on mathematical models that are clearly meeting their objectives of accurate, fair and equitable. In our analysis based on recent sales by price range, we determined that 1.25 is the best multiplier to determine the value of a property using the FCV model. A restricted supply of homes and consistent demand for homes in the lower price ranges are causing strong appreciation gains, while higher priced homes are seeing limited price gains due to an increase in supply. According to the model, here are the expected valuations based on price ranges equally divided by volume:
Leaving the FCV model and looking at price per square foot models, the two biggest challenges are properly defining the geography which identifies similar properties and having a large enough sample size to garner meaningful results. Regardless of how strong your data set is, bad data always exists and incorrect data will garner incorrect results.
In conclusion, coming up with a good AVM is a daunting challenge, but to be honest, it’s kind of fun. I believe the greatest value of an AVM is that it creates an interesting tool for investors wanting to identify undervalued assets, but for the individual homeowner they simply provide water cooler conversation, some boasting, some complaining and some agreeing. Again, the best way to know the true value of a home is to consult with a real estate professional.