I have chosen to begin my 2019 oeuvre with an inflammatory question: Can we predict where real estate prices may be heading over the short term?
There are two, related reasons for investigating this. First, should the overall economy continue to expand through July 2019, it will represent the longest economic expansion in this country in its 242-year history. Ever. (Regardless of which political figures take the credit, it is a fact.)
In addition, we know that real estate prices in Greater Fort Lauderdale have been increasing briskly since mid-2011. As of last September 30th valuations have advanced, on average, by 81 percent from that trough.
For how much longer will this uptrend continue? And are there some early warning markers out there that might give us an inkling of an imminent change in the trendline?
To begin to analyze this, I downloaded metro Fort Lauderdale data from the Federal Housing Finance Agency, which go back to 1975. (It is a good thing I didn’t have any questions about the data. As of my writing, the Federal government is shut down. Sadly, statisticians in the Housing Department are not considered “essential Federal employees”.)
From this data, we may easily calculate relationships in prices, comparing what we knew as of a certain date with what then happened over the near-term.
Bear in mind that you cannot predict precisely what is coming. Rather, I think there is some value in determining whether established trends might continue, as well as identifying turning points on the major trend. This is especially the case for people who do not have a firm date by which they must buy or sell.
In the table included here, you can see how closely pricing in future quarters relates to the current pricing. Perhaps not surprisingly, the closer you are to the present, the more accurate the prediction you can make. As more time elapses, the prediction becomes less robust, holding everything else constant.
The results bear similarity to “technical analysis” of security prices, in that an established trend will tend to continue. If the pricing trend is up, near-term future pricing tends to be higher, and vice versa.
To summarize Stage One of the modeling:
1. If housing values are INCREASING, they will tend to continue to INCREASE over the next year or so (but at a slower rate); AND
2. If housing values are DECREASING, they will tend to continue to DECREASE over the forthcoming 12 months (but at a slower rate); UNLESS
3. “Something happens”.
In my next column, I will take a look at whether there is a “something” or “somethings” that can be shown statistically and mathematically to suggest an imminent inflection in the trendline.






