“Paired Sales are a joke” …So some folks say.
Okay, I agree that paired sales are not the be-all and end-all of adjustments, and that there are other methods for supporting adjustments, but the paired sales method really can be valuable.
Yes, paired sales are much more accurate in a textbook or in a state board’s newsletter than in an actual real-world environment. It’s wonderful if you can say “these two properties are exactly alike, except this one has a three-car garage, and that one has a two-car garage – so that means a third bay is worth $10,000”. The drawback, of course, is that it’s often difficult to find two properties similar enough (let alone, exact!) to even begin to isolate just one single characteristic for establishing adjustments. So, should we just give up on paired sales?
I think there are a few misunderstandings surrounding paired sales. Paired sales are not designed to arrive at an exact rate of adjustment to be used for each and every property you come across in each and every assignment. Simply because you calculate a $43.7982 per sq.ft. adjustment rate based on two sales doesn’t mean you’re going to use that exact rate for every assignment. Perhaps that support is only valid in a certain subdivision, or to new construction, or in a certain market area, or for certain age properties. If you determine a third garage bay is worth $15,000 for a certain segment of properties, will it also be worth $15,000 in so-called higher-end houses? Maybe, maybe not. Perhaps the $15,000 represents a percentage, not a flat dollar amount. Just like anything else, paired sales require analysis, not just calculations.
And we won’t ever find the perfect textbook example of identical-except-for-one-characteristic paired sales. But is this reason to abandon the process? Paired sales will undoubtedly require adjustments in order to isolate a single characteristic. For example, maybe one is on a larger site, but you can reasonably adjust for site size differences, and then isolate the sq.ft. rate. Or, perhaps one property has slightly different GLA, so you qualitatively know that the difference in sale prices isn’t entirely attributed to the two-versus-three bay garages, but you can still draw a conclusion regarding if and how much value a third garage bay adds still taking into account that their GLA vary slightly.
Additionally, one paired sale is not the end of the process. Over time, we should be collecting, analyzing and storing the studies so that we can develop ranges of market-supported adjustments. Maybe our data indicates a garage bay is worth between 5% and 15% of the value for a particular property type in a particular area. So, where in that range does the appropriate adjustment fall for a specific assignment? …You are the human market expert qualified to develop the reasoning to support the specific adjustment you use in a report, which will most likely be based on the range your research indicates.
Paired sales will not show an exact number or percentage. So, what do they do? What use are they? First, they establish that there is (or is not) a value contribution associated with a specific characteristic. Second, they provide a basis for the adjustments that we end up making. A basis does not mean “a number to blindly follow without thinking about it”! A basis means our adjustments are logically grounded in real analysis, and not simply pulled out of thin air.
Of course, there are other methods of supporting adjustments: regression, market interviews and similar, which can be especially helpful in rural and non-homogeneous markets. Yes, only in textbooks are paired sales perfect. But to simply write them off because they are imperfect in the real world ignores the real evidence and support they can provide.