3 Secrets to reading housing data for the real estate investor: Understanding seasonally-adjusted data

Note: this is part 1 in a 3-part series entitled “3 Secrets to reading housing data for the real estate investor”.   Part 2 was about understanding margin of error. Part 3 was about Looking beyond the trend.

Part 1 – Understanding Seasonally Adjusted Data

If you’ve ever read market reports, economic data, housing statistics, etc, you’ve probably experienced that strange sense of wondering whether the numbers in front of you are really all that  meaningful or significant. Sure, they may seem impressive, but, is up really up, and is a trend really a trend? Are we really seeing what we think we’re seeing? Do our figures represent the whole picture, or are they distracting us from another truth?  As one Aaron Levenstein once said, “Statistics are like bikinis.  What they reveal is suggestive, but what they conceal is vital.”

In order to help us wade through these broad existentialist dilemmas, Maria Sencovici writes a great piece on 85 Broads regarding the ins and outs of reading housing data. I’ve taken her article and expanded it with my own thoughts.  So here then is a short mini-series covering the 3 most important things that any real estate investor must understand when trying to make sense out of housing, economic, real estate, or any other large-scale data.

We’ll start today with part 1.

Part 1 – Understanding how to interpret seasonally-adjusted data

The real estate industry is highly seasonal. For example, most buyers buy in the fall, most renters lease in the summer, and so on. In order to compare data across contiguous time periods (e.g. comparing spring to summer, and summer to fall) researchers “seasonally adjust” data to make it more useful and to allow readers to draw more intuitive comparisons.

Reading that SA housing starts are up by 20%, for example, doesn’t mean that starts themselves are up by that much; rather that they beat the expectations of the smoothed out numbers we would have seen if we ignored seasonal influences.

And, looking at it from another angle, reading that NSA housing starts are down by 20%, you might think about the time of the year and ask yourself what you would expect, all else being equal. For example, if you’re now in colder weather, it makes sense that you would expect fewer housing starts, and the question then is whether the 20% decrease is higher or lower than you would have expected.

The trick lies in knowing which kind of data you are reading — you will often see seasonally adjusted data reported as “SA”, and not seasonally adjusted data as “NSA” — and in how to read it.