Excessive GLA and improvements

[stangtime]

Well done. I play with/ work with the graphs in Excel 2007 but not to the level that you have perfected.

Dan

 

[vanguard]

The big issue is, if you are appraising to market value, would a typical homebuyer use your formula when they purchase this property or are you just dressing up your analysis to impress the client. Someone may just want to purchase it because they like it and see its' potential.

 

[Lisle]

Interesting thread. However, at what point do you say, the value for this property cannot be reliably evaluated by sales comparison?

 

[Metamorphic]

Well done. I play with/ work with the graphs in Excel 2007 but not to the level that you have perfected.

Dan

Its really not that big a stretch. Anybody who's generally familiar with Excel I could teach it to in a couple hours.

 

[hastalavista]

Interesting thread. However, at what point do you say, the value for this property cannot be reliably evaluated by sales comparison?

(my bold)

For GSE work, that point occurs right before you withdraw from the assignment, since your question implies that the sales comparables are not meaningful or reliable.

The graph and its supporting analysis does not explain why market participants make their purchase-decisions. In other words, as Robert A. points out, it is doubtful that the typical buyer actually goes through such an exercise. But just because the typical buyer does not go through such an exercise doesn't mean the data isn't useful to predicting how a typical buyer (the market) will react given "size of the home" and "price of the home".

The graph and its supporting analysis simply presents what the market reaction is to the two variables; in this case, gross living area and price. I can theorize why the market pays more for additional GLA and I have fairly good empirical evidence to support that theory. But to value the property, what I need to do is develop a model of what the reactions of the market are to different components, and then apply that market reaction to the properties I'm comparing to my subject to account for the differences.

I would not rely (and no one participating in this thread is suggesting) on strictly a statistical analysis to prove a point. I would (and I do) rely on statistical analysis to provide support for my adjustments. In this example, the graph provides a visual of the relationship between two variables which is much more effectively presented in this form than in a table or narrative. I can take this graph and add it to my argument that:
Above a certain range, there is insufficient market evidence of demand for homes in excess of X square-feet; therefore, I've applied the following GLA adjustment scheme to this assignment.....

Statistics is just a tool to use. With any tool, there is a danger of it being misused (intentionally or unintentionally). Relying only on statistical analysis is what AVMs do. Using statistical analysis in conjunction with all the other tools appraisers regularly use adds persuasiveness to the appraiser's supporting arguments and credibility to the overall value opinion.

My two-cents! :new_smile-l:

 

[Metamorphic]

Interesting thread. However, at what point do you say, the value for this property cannot be reliably evaluated by sales comparison?

To what Dennis said I'd add:

If the property cannot be reliably evaluated by SCM, your appraisal problem, as a practical mater, would become much more about characterizing the HBU. As soon as you assert that its not possible to value something based on a specific kind of market sale, it seriously undermines the idea that the property is marketable for the use for which you were trying to find sales data.

Perhaps the best way to illustrate that is to flip it around.
"I'm trying to value this 25,000SF warehouse and I cant find any sales of warehouses like this for use as single family residences."

_____ "Yeah, no feces." :Eyecrazy:

"I guess I'll have to dump the SCM and use the income approach. Call the management company and see if they have any big warehouses rented to a couple with 2.3 kids and a dog."

_____" m2:m2: "

In my first post I proposed the idea that maybe some other HBU was more appropriate, but the OP shut the door on that idea, so I have to assume they are convinced (for good reason) that the HBU is Single Family. If I were reviewing this eventual report, I'd want to see a serious discussion of why this monster house cant be a Lodge, half-way house, boarding house, hostel, corporate retreat, BnB, demo'd and turned into a buildable lot, subdivided into a quad-plex, etc.

 

[DMZwerg]

The big issue is, if you are appraising to market value, would a typical homebuyer use your formula when they purchase this property or are you just dressing up your analysis to impress the client.

Whether a homeowner would use a particular formula or not is often not relevant. Just because a landlord uses a formula to determine what he should charge for rent does not mean a potential tenant will, rather the tenant will likely look at market rents already existing then decide if it fits his specific needs. One could well be looking at things in general and another at specific things yet still come up with the same answer.

As an appraiser I can look at trends to get a general feel for things, but those trends are based at first on actual transactions. As with any statistics there is a percent chance they are utter hogwash and that is where the more "standard" appraisal approaches to value come into play. Anyone who relies on a single formula or graph without verifying and supporting the opinion with additional sound appraisal practice deserves to get creamed.

 

[Lisle]

The graph and its supporting analysis simply presents what the market reaction is to the two variables; in this case, gross living area and price.

How has the reaction to the variable of GLA been extracted/isolated? In higher end properties we many times are given several high value items to evaluate. For example, waterfront, waterview, large acreage, etc...

Also on an unrelated matter, While using the forum today I am having to parse through alot of code when replying. I think it's VB and HTML code. Sorry to put the question to you here, but along with not being able to post a new thread some other controls are also not operating correctly.

 

[Lisle]

To what Dennis said I'd add:

If the property cannot be reliably evaluated by SCM, your appraisal problem, as a practical mater, would become much more about characterizing the HBU. As soon as you assert that its not possible to value something based on a specific kind of market sale, it seriously undermines the idea that the property is marketable for the use for which you were trying to find sales data.

Perhaps the best way to illustrate that is to flip it around.
"I'm trying to value this 25,000SF warehouse and I cant find any sales of warehouses like this for use as single family residences."

_____ "Yeah, no feces." :Eyecrazy:

"I guess I'll have to dump the SCM and use the income approach. Call the management company and see if they have any big warehouses rented to a couple with 2.3 kids and a dog."

_____" m2:m2: "

In my first post I proposed the idea that maybe some other HBU was more appropriate, but the OP shut the door on that idea, so I have to assume they are convinced (for good reason) that the HBU is Single Family. If I were reviewing this eventual report, I'd want to see a serious discussion of why this monster house cant be a Lodge, half-way house, boarding house, hostel, corporate retreat, BnB, demo'd and turned into a buildable lot, subdivided into a quad-plex, etc.

Thanks for the reply Meta. Well put, its a keeper. I would think some Cert General input might be in order for determination of commercial HBU.

 

[hastalavista]

How has the reaction to the variable of GLA been extracted/isolated? In higher end properties we many times are given several high value items to evaluate. For example, waterfront, waterview, large acreage, etc...

No one is arguing that there are not other factors that influence price. Looking at the graph, at the 2,000sf mark, there is a wide range of values; this implies that there are significant factors at work that influence values for 2,000sf (one of them could be market conditions).
Nonetheless, is it reasonable to draw a conclusion that there a relationship between size and value and that, given the data, the subject is outside the data sample. Further, is it reasonable to argue, based on the data, that there is a diminishing return of value based on incremental increases of GLA after a particular point.
At point $X, the typical buyer will not purchase in this neighborhood; the typical buyer will go to another neighborhood where $X is transacted. And, that is the purpose of the analysis, no? To determine how much value the market can support for GLA differences before the market determines the GLA is over-improved and does not pay any more. :new_smile-l:

That is the benefit of the graph (again): to provide a visual display of the relationship of two variables. The value of the appraiser is to take this data and disaggregate it down to the the specific assignment and make adjustments or weightings that are market-supported.

Do appraisers not collect a similar-type data set every time they complete the 1004mc? In the 1004mc example, the data set is typically smaller because the data is homogeneous (similar to the subject).

I've attached a graph I've completed using data for a property I'm appraising in Santa Clara County, CA; I'm not going to use this graph but I have the data for other analyses I'm completing. I've defined a competitive market area, and have run all MLS sales, detached SFRs, on lots of 6k to 20k. I've circled where the data clusters. Outside of this cluster of prices, it would be an extraordinary event for a home to sell.
This display illustrates (to the reader and to the appraiser) where this market typically prices homes based on GLA. If one were to value a home outside of the cluster, one would (presumably) need a very good reason to do so; and that reason probably isn't additional GLA. Therefore, analyses like this help the reader (and appraiser) understand how the market is functioning in a one-look display.

And, yes, from this, I can extract (based on the market) a point or range, where there is not enough credible evidence that additional GLA will provide a discernible incremental increase in market value. That is the point of over-improvement for GLA where no additional value is awarded in the market.

Lisle, I think you make (or imply) two significant points
(a) the data can be dorked to show something which really isn't, and
(b) one must be careful about reading too much into the data as there may be other variables.

I'm assuming that the appraiser is not trying to BS the reader.
As to the second point, I agree with DMZ

As with any statistics there is a percent chance they are utter hogwash and that is where the more "standard" appraisal approaches to value come into play. Anyone who relies on a single formula or graph without verifying and supporting the opinion with additional sound appraisal practice deserves to get creamed.