“There are three kinds of lies: lies, damned lies, and statistics.”
This quote, usually attributed to Mark Twain, is one of my favorites. I learned it early on in my economics education, and have repeated it often. It humorously highlights that statistics are just numbers. To have meaning these numbers must be understood in context, with an understanding of underlying relationships between data. Without context, statistics are prone to be misinterpreted and misconstrued.
The housing market has been on a roller coaster for more than a decade. In my neck of the woods, prices have skyrocketed during the past several years. For appraisers, these changing market conditions have necessitated adjustments for time. How those adjustments relate to the market statistics reported in the news media is something that I’ve had to explain many times. The question is often posed as follows: “If prices have gone up by 15% during the past year, then why didn’t you adjust at a 15% rate for time?” Of course, the simple answer is that the analysis of my comparable sales data did not support such an adjustment. But why?
The statistic often quoted for real estate prices is the percentage change in median sales price over a period of time. Within this statistic, there are a myriad of factors which change over time and which impact the result. Let me illustrate using some data gathered from North Texas Real Estate Information Systems, which is the MLS provider in my market. The chart below compares median home prices in Collin County, TX.
| Date | Median Sales Price | % Change |
| July 2014 | $260,000 | |
| July 2017 | $325,000 | +25% |
Extra! Extra! Read all about it! Collin County real estate prices have increased 25% during the past 3 years! Well, this is true in the sense that it is correctly calculated, but what does it really tell us about the Collin County real estate market?
A little bit of digging revealed that the median GLA of a home sold in Collin County in July of 2014 was 2241sf. Three years later, that median had risen to 2444sf. Given that bit of information, I have plotted the per unit pricing below for comparison.
| Date | Median Sales Price/SF | % Change |
| July 2014 | $116/sf | |
| July 2017 | $133/sf | +15% |
Hold the presses! The story changed significantly with the inclusion of just one additional variable. There are many others of course. Trends in the market over time may impact the composition of the sample data. Take new construction for example. In July 2014, new construction accounted for 10% of sales, whereas by July 2017, new construction sales accounted for 15% of the sales reported to the MLS. The greater proportion of new homes had a significant impact on the reported magnitude of changing prices.
In general, the analysis of aggregate economic data such as median sales prices, should be very similar to the way you analyze a prior sale. Just looking at the prior sales price alone doesn’t tell the full story. Questions need to be asked about what sort of changes might have been made to the property since the prior sale.
The same logic applies with aggregate data. If expressed as a formula, it would look like this:
Current median price level
– Prior median price level
– Aggregate depreciation during interim period
+ Aggregate contributory value of investment/improvement during interim period
= Actual supply/demand related difference in price
÷ Prior median price level
✕ 100%
= Percentage change in price level due to market forces
The prior price level less aggregate depreciation, plus aggregate investment/improvements, is what needs to be compared to current price levels in order to determine the actual change in market prices. Is your subject in an older area with an increasing trend toward remodeling? Chances are the aggregate depreciation is well below the aggregate investment in that market. In that case the magnitude of an increasing market may be well below what is reported.
You may never get a dissatisfied homeowner or sales agent to understand this concept, but being able to explain these statistics may make the difference between keeping a good client, or having to find a new one.
