GLA in Sales Comparison Approach

[Mike Boyd]

I'm with Doug W. and George D. From time to time I will complete the market approach first except for the SF adjustment, if any. That is your MATCHED PAIRS. Once you have adjusted everything out, they are matched. You can do that for each adjustment. Sometimes I will take inflation into account if I have not done the matched pairs analysis for awhile. Most of the time I can do it in my head. I read something years ago that said you should not adjust at more than 25% of the PSF replacement cost for average quality homes and maybe 33 to 35% for good quality. Excellent quality was left open because there are so many quality factors to consider.

 

[Joyce Potts]

Interesting answers. I agree with some of many posts and totally disagree with some on others.

But the post I think I agree with the most is George Dodd. That's pretty much where I stand on the matter, right, wrong or indifferent.

And where is Mechanicsville in relation to Annandale?

 

[Fred]

I'm reading this as the opperative word is "SUPPORT" their adjustments - not "how do you calculate" your adjustments

1. If the market provides enough sales to “calculate” the adjustment, then that is the only “supported” way. Explaining is not necessarily supporting.
2. If the market doesn’t provide enough data (for example, the 6,000 sf white-elephant house in a 2,000 sf neighborhood) then one must use a principles-driven adjustment – in this case diminishing returns. By showing a sound application of the principle, the adjustment is partially supported. An appraiserl might have a file with white-elephant sales studies and could use the rates in those studies as “comps,” again partially supporting the adjustment.
3. It is possible to say that the adjustments are pro forma to help the reader visualize how the range of the comps narrows around the “center.”
4. If you use ranking methods, there are no size adjustments.

In the first situation, adequate data, how can the size adjustment be anything other than what the sales in the neighborhood or the relied-on subset of sales indicate? The sizes and prices themselves show degree to which changes in size correspond to changes in price. There are a couple of ways of yanking the number out of the sales. To answer the original post, that’s I use. So naturally, I don’t agree with this.

It is a long process that comes with many years of looking at sales data and subject properties
You do this over and over and over and over again. At some point you will recognize that there are ratios and ranges for properties in your market.

For me, it is a relatively simple process that takes seconds.

value is not a USPAP compliance matter

To the contrary, the document itself has numerous “must” requirements that call for “credible results” or a “credible appraisal” (i.e. credible opinion of value). The only assignments I have ever had in which USPAP requirements per se became a significant were because someone suspected that there was a lack of “credible results.”

 

[George W Dodd]

And where is Mechanicsville in relation to Annandale?

Annandale is about 80 miles north of Mechanicsville, just about to DC. Mechanicsville is just north/northeast of Richmond.

 

[Peter LeQuire]

My take on this is that there are several acceptable ways to develop supported GTA adjustments: extraction is certainly one, and the one I favor as being relatively simple to apply and explain. Regression analysis certainly is recognized as defensible. I've used graphic analysis on occasion. Admittedly, there are times when there is so little similarity between the victim property and sold properties that ranking them and making no GLA adjustments - as Steven points out - has been the only way to wander in on a value opinion. Matched pairs work if you've got pairs that are matched.

The point is to support what is done.