Adjusted Sale Price Per Square Foot (asp/sf)

[BubbaP]

I realize that this same topic was addressed in at least one other AF discussion a couple years ago (https://appraisersforum.com/forums/threads/adjusted-sales-price-price-per-sqft.216540/ back in April-May of 2017); however, IMO the original poster's primary question in that discussion was not definitively/clearly answered, and no supporting documentation was presented/referenced. Therefore, because I need a definitive answer, and especially supporting documentation, I have decided to re-ask the same question, which is:

In order to determine the Adjusted Sale Price per Square Foot (ASP/SF) of each comp in a residential appraisal or valuation, do you divide the adjusted sale price (ASP) of each comp by each comp's or the subject's square footage?

I realize that ASP/SF is not the standard unit of value used in residential appraisals (as mentioned in the original discussion) and that it is not used on the standard URAR form, and therefore that this question may appear to have little to no relevance. However, in my current position as a tax rep, I often see assessors use ASP/SF in defense of their residential property values, and IMO, it is erroneously calculated by vast majority of them. Therefore, this topic is obviously very relevant and important to me.

I'm of the opinion, like the original poster in the aforementioned discussion, that the subject's, and not the comp's, square footage should be used to determine the ASP/SF of each comp. IMO after adjusting the sale prices of each comp in order to make it comparable (approximately equivalent) to the subject's probable sale price, that this is the only thing that makes sense.

I have discussed this topic with a couple of my most trusted/experienced appraiser friends, and though both initially disagreed with my position, after further explanation (by me) and thought (by them), both ultimately agreed with my position. I'm therefore ~99% convinced that my position is correct. However, I've not been able to find any documented proof, either for or against my position. I would very much appreciate documented proof, either for or against my position.

Thanks in advance for your consideration and help!

 

[VolcanoLvr]

Your question: "In order to determine the Adjusted Sale Price per Square Foot (ASP/SF) of each comp in a residential appraisal or valuation, do you divide the adjusted sale price (ASP) of each comp by each comp's or the subject's square footage?"

Is revealed on the GSE (and GP) forms. It's at the top of the grid for each COMPARABLE.

That figure is the Comp Sale Price divided by the GLA of that Comp. (Does not include any below grade space.) This is NOT the "Adjusted Price Per Sq. Foot" which is the basis of your question.

I'm not following your logic of using only the Subject property as the divisor for this problem, when multiple properties are involved - which would have differing $ per s/f figures.

What you are discussing is likely a Mass Appraisal algorithm method whereby the Assessors are trying to equalize cost per s/f for taxation purposes across homes in specific neighborhoods.

 

[Restrain]

The Adj Price /SF for an analysis would be based on each comps square footage but this is more often found when doing a multi-family property. It is often a part of the overall analysis including Price per Room, Price Per Bedroom, GRM per Unit, etc. Then you would analyze each comp and try to figure out what the probable Adjusted Price /SF for the subject would be and multiply it by the subject square footage to reach one of the indicators of value. For the SFR, you are not so interested in the Adjusted Price Per Square Foot but the adjusted sales prices of the comparables after ALL adjustments. A group of adjusted sales prices may be, for example, $175,000, $180,000, $182,000, and $185,000. You would weigh the sales, looking at each of the sales and saying 'Sale 1 had the fewest number of adjustments, but Sale 3 had the lowest dollar of adjustments, and was most similar.

When you are doing a SFR, which I think you are looking at, you can make the adjustments for other items than size (garages, site, etc), then analyze the difference in sizes of the comparables and adjusted sales price excluding the adjustment for size to reach the Adjustment for square footage, which is adjusted against the Subject's square footage.

I hope this helps.

 

[Terrel L. Shields]

Depends on how you do your math...
If you have a per SF adjustment estimated from the comps, then you apply it to the subject SF.
But no I would not use the subject the way you describe.
Say you have one comp that you determine by paired sales, added $10,000 for crown molding (as an example) and the house is 2,000 SFD or $5/SF to the price. And another comp is 1800 SF and added $9,000 to the value of the house...OK the SF adjustment is $5. So if the subject is 2400 SF. The adjustment is not $10K or $9K, rather is $5/SF or say, $12,000. Now crown molding is a poor example but you could apply it to age / condition, etc. And in commercial work, I often do that.

 

[Artemis Fowl]

I'm of the opinion, like the original poster in the aforementioned discussion, that the subject's, and not the comp's, square footage should be used to determine the ASP/SF of each comp. IMO after adjusting the sale prices of each comp in order to make it comparable (approximately equivalent) to the subject's probable sale price, that this is the only thing that makes sense.

Totally lost me here. What exactly is this process going to illustrate about the market? What is revealed by doing it this way? How do those answers provide insight into the subjects' ranking in its' market?

 

[NachoPerito]

You are correct OP.

If you already have made an adjustment for size then you divide the adjusted sale price into the comp square footage that would be essentially making the same adjustment twice. Not proper.

Example:
Comp 1 = 3000 SF , $450,000
Subject 2000 SF , $360,000

You adjust comp 1 downwards $60,000 for size = adjusted price of $390,000.

You conclude the adjusted price equals $390,000/3000 = $130/SF? No that doesn't make sense....

Of course, I don't get why you are doing what you are doing. I have never done that. But structurally it makes sense.

 

[BubbaP]

Judging by some of the responses so far, which IMO go off into a few unnecessary tangents, I’m not sure I did a good job of stating the question/issue. I think this might be better illustrated with a simple example. Suppose we have the following appraisal where all 3 comps are essentially equal to the subject (same builder, subdivision, construction quality, age, condition, location, view, bed/bath count, etc...), except for differing GLAs:

Subject: 2000sf
Comp 1: 1950sf; Sold for $300,000; $153.85/SF
Comp 2: 1800sf; Sold for $285,000; $158.33/SF
Comp 3: 2200sf; Sold for $310,000; $140.91/SF

The above $/SFs are sale price per square foots (SP/SF), not adjusted sale price per square foots (ASP/SF). This is straight forward and is not debatable.

Now apply GLA adjustments to the comps @ $60/SF:
Comp 1: $300,000 + (50SF x $60) = $303,000
Comp 2: $285,000 + (200SF x $60) = $297,000
Comp 3: $310,000 - (200SF x $60) = $298,000

This results in a nice, tight ASP range of $297,000 to $303,000. At this point I’m sure that most of us would simply conclude the subject’s value at around $300,000 (probably giving most weight to comp 1) and be done with it. I know I would. However, some/many tax assessors (and boards) prefer to convey values in terms of $/SF, so they convert each comp’s ASP into an ASP/SF. The question is: what is the proper way to do this? Most (but not all) assessors do so by dividing each comp’s ASP by its own SF, as follows:

Comp 1: $303,000 / 1950SF = $155.00/SF
Comp 2: $297,000 / 1800SF = $165.00/SF
Comp 3: $298,000 / 2200SF = $135.00/SF

This results in a wide ASP/SF range of $135.00/SF to $165.00/SF, which is considerably wider than the SP/SF range of $140.91/SF to $158.33/SF. IMO this doesn’t make sense because the adjusted value range (whether using ASP or ASP/SF) should always be tighter than the unadjusted value range. Alternatively, dividing each comp’s ASP by the subject’s SF results in the following:

Comp 1: $303,000 / 2000SF = $151.50/SF
Comp 2: $297,000 / 2000SF = $148.50/SF
Comp 3: $298,000 / 2000SF = $149.00/SF

This results in a nice, tight ASP/SF range of $148.50/SF to $151.50/SF, which is considerably tighter than the SP/SF range of $140.91/SF to $158.33/SF, as it should be IMO.

I hope this clarifies the question/issue. Whether I’m correct or not, I would very much appreciate documented proof. I would think this topic would be covered in some appraisal text book, or that some highly designated appraiser guru would have published an article on this. If not, then IMO somebody should…

Thanks again!

 

[NachoPerito]

Judging by some of the responses so far, which IMO go off into a few unnecessary tangents, I’m not sure I did a good job of stating the question/issue. I think this might be better illustrated with a simple example. Suppose we have the following appraisal where all 3 comps are essentially equal to the subject (same builder, subdivision, construction quality, age, condition, location, view, bed/bath count, etc...), except for differing GLAs:

Subject: 2000sf
Comp 1: 1950sf; Sold for $300,000; $153.85/SF
Comp 2: 1800sf; Sold for $285,000; $158.33/SF
Comp 3: 2200sf; Sold for $310,000; $140.91/SF

The above $/SFs are sale price per square foots (SP/SF), not adjusted sale price per square foots (ASP/SF). This is straight forward and is not debatable.

Now apply GLA adjustments to the comps @ $60/SF:
Comp 1: $300,000 + (50SF x $60) = $303,000
Comp 2: $285,000 + (200SF x $60) = $297,000
Comp 3: $310,000 - (200SF x $60) = $298,000

This results in a nice, tight ASP range of $297,000 to $303,000. At this point I’m sure that most of us would simply conclude the subject’s value at around $300,000 (probably giving most weight to comp 1) and be done with it. I know I would. However, some/many tax assessors (and boards) prefer to convey values in terms of $/SF, so they convert each comp’s ASP into an ASP/SF. The question is: what is the proper way to do this? Most (but not all) assessors do so by dividing each comp’s ASP by its own SF, as follows:

Comp 1: $303,000 / 1950SF = $155.00/SF
Comp 2: $297,000 / 1800SF = $165.00/SF
Comp 3: $298,000 / 2200SF = $135.00/SF

This results in a wide ASP/SF range of $135.00/SF to $165.00/SF, which is considerably wider than the SP/SF range of $140.91/SF to $158.33/SF. IMO this doesn’t make sense because the adjusted value range (whether using ASP or ASP/SF) should always be tighter than the unadjusted value range. Alternatively, dividing each comp’s ASP by the subject’s SF results in the following:

Comp 1: $303,000 / 2000SF = $151.50/SF
Comp 2: $297,000 / 2000SF = $148.50/SF
Comp 3: $298,000 / 2000SF = $149.00/SF

This results in a nice, tight ASP/SF range of $148.50/SF to $151.50/SF, which is considerably tighter than the SP/SF range of $140.91/SF to $158.33/SF, as it should be IMO.

I hope this clarifies the question/issue. Whether I’m correct or not, I would very much appreciate documented proof. I would think this topic would be covered in some appraisal text book, or that some highly designated appraiser guru would have published an article on this. If not, then IMO somebody should…

Thanks again!

Read my comment above. Your range gets wider because you are essentially adjusting for size twice.

 

[BubbaP]

Read my comment above. Your range gets wider because you are essentially adjusting for size twice.

OK, then it sounds like you agree with me, right? Any published/documented proof of this???

 

[Restrain]

There is a significant reason that appraisers generally don’t apply the Gross Price/SF. It doesn’t take into consideration all the differences between the subject and the sales. Real estate agents tend to do this as do AVMs but you end up with wide ranges,