Functional vs External

[Tom Hildebrandt]

We have an ethical obligation to be competent under USPAP.

To be competent, USPAP rquires that we apply judgement to defining the appraisal problem, and correctly research the markets and apply what techniques and methodologies are pertinent to helping solve the problem in a credible manner.

USPAP does not say anything at all about bracketting properties or how to solve the problems except on a conceptual basis. The "how too" is from texts, seminars, etc. Just because you have no direct data for an adjustment does not mean that you can not make one based on logic and reason.

What Terrel has done is what needs to be done. He identified the underlying issue which makes this market valuation problem complex, that perhaps the home is overbuilt for the market. He has provided a possible explanation and has presented a possible solution based on his analysis of the data. Just because you can not "prove" an opinion (or an adjustment) is the only correct one does not mean it is any less valid, so long as the process is supported by evidence and logic. The support for the obscolescence adjustment can be seen in the difference between the two markets, and can be measured. Whether it is correct to take the adjustment is the issue and a matter of professional judgement.

In this case, the lender needs to know the risk asociated with this property. I would provide two values, one showing the market without the possible obsolescence, and the other with the appraisers take on the obscolesence. In the reconciliation, say that there is insufficient data to allow the appraiser to form a definitive position on where the market would go, but the data does suggest a possible range. The appraiser can make a judgement as to what he would like but advise the client that he could make any determination he so chooses in consideration of the risk.

The appraiser would have donme his job, andd put the burden on judging the risk squarely where it belongs, the bank.

Regards

Tom Hildebrandt GAA

 

[jtrotta]

Austin;

Your statement; "I would not put any faith in interviews with Realtors or any body else. If Realtors knew anything we wouldn't need appraiser's."

*Perhaps your skill in extracting information needs improvement, as I do not see it that way, thats like a carpenter throwing away needed six penny galvanized nails for a standard six penny nail (your analogy-with doctors). By not interviewing the supply of information around you, you only limit your knowledge of your own marketplace. Therefore, by not conducting any interviews with Builders; Lawyers; Realtors; Buyers & Sellers; you limit your knowledge of the market; market trends and what happens within your own industry.

Therefore, in my opinion Austin, your dead wrong :!:

 

[George Hatch]

.

I got $.02, I got $.02......


Terrel's original question was whether to call it functional or external. I think the argument can be made both ways. I personally favor the functional label because of the apparent violation of highest and best use. Had the design and scope of the improvements been more in line with the HABU, there would be no loss. The design could have been changed, but the location remains the same. Thus, "functional obsolescence due to the failure to recover the full cost of construction in the market". The fact that there are a few other freaks in the local area that have not sold merely shows there are a few other freaks in the area.


As for the solution to the valuation problem, I consider location adjustments to be the least desirable adjustment to make. The Wally-world homes are not comparable because of the different location with dissimilar economics, and is demonstrable because there are some sales of those homes over there. Those homes are not directly comparable to Terr's subject, despite the similarities in size and design. Better to find similar overbuilts in more similar locales. Go outside the county to find an economically comparable area, and see if there are any sales of freak homes. Those are your comparables. Using the Wally-world properties as comparables, without significant adjustment for location, is almost as silly as taking them from Dallas or Palm Beach.

Oh yeah, and exposure time will probably be an issue in this assignment. An estimated exposure time that far exceeds the norm will tell your reader that the marketability of the property within that market segment is poor. That would kill a highly leveraged deal all by itself with most lenders, regardless of the value estimate. "Estimated exposure time is in excess of 12 months." Terrel might even be asked to provide an additional value estimate based on a reduced exposure time.

Like I said, it was $.02.

George Hatch

 

[KD247]

Austin,

Sorry, I don't follow your logic. Let me restate it and see if I've got it right. Are you saying that because the word "probability" has a specific meaning within the field of statistics, that the word can't be used in USPAP to simply mean likelihood?

Businesspeople are continually making decisions based on probability. Like, "What's the probability that employees will leave if we buy that company". Only some percentage of these decisions can appropriately utilize statistical methods.

So, is your conclusion that because USPAP uses the word probability there is an implication that a statistical approach *must* be used?

Your analogy regarding the existence of unique properties confuses me too. As an appraiser, I am primarily concerned with the market's reaction to (most probable sales price) the subject property. I am not overly concerned with the probability of similar properties existing in other areas.

With the lack of data in this case, I think that construction costs for similar new properties provide a good idea of what someone is willing to pay for a similar property. Of course, the motivations of these builders must be compared with motivations of typical buyers.

Also, it is not unusual for appraisers to look outside of the immediate market area. It's easy to imagine that the typical buyer for ultra-high value estate properties might even compare properties in different countries.

If our job were entirely data analysis, we would have been gone a long time ago. I don't go along with the analogy that appraisers are like doctors and shouldn't be asking the patient for a diagnosis. I think appraisers are more like a member of an audience trying to predict which way a dancer will go next. I would spend a good deal of time looking at dancers, but I would also put a lot of energy into talking with the dancers, choreographers, stage hands, and other members of the audience.

You ask us to put this case in perspective with the amount of work required to hone down a $3,000 adjustment, compared to, in this case, the uncertainty of supporting a $100,000 adjustment. Well, here's another perspective. In Santa Barbara, homes regularly sell for $5-$10 million, with very few comparables and a low likelihood that some will be even vaguely similar. Yet, appraisers routinely make adjustments of up to $1-$2 million and come up with their best estimate of value.

This is business as usual in many parts of the country and appears to adequately comply with lending requirements. Are you saying that if the typical buyer is willing to spend an extra $700,000 for an oak-lined cobblestone driveway, that we appraisers should stop the buyer's gross negligence, based on a spike in our trend line?

We probably have different viewpoints because we appraise in different markets. You have found a successful solution to most of your appraisal problems by applying statistical analysis. I have most often found successful solutions to my appraisal problems by putting myself in the shoes of the typical buyer.

A theoretical "absolutely typical" buyer would probably look at this property and say "I'd pay more than for the smaller homes in this area, but less than I would for the similar homes in a better area, I'm not going to pay more than what I could build it for, and furthermore I'd want a pretty big discount for this oddball." Maybe that's as precise as the analysis can get for this property.

To use your lender's question... “Can you Mr. appraiser tell me with a high degree of probability what the subject property will sell for within x time period with sufficient accuracy to protect my investment?”, my answer would be "I can be relatively confident that the property would sell at some point within an admittedly wide value range, and I can select a point within that range as my best estimate of market value. I can also provide an estimate of marketing time, but that will also be expressed as a fairly wide range. It's your job, as the lender to decide whether your investment is safe at this level of accuracy."

Ok, Koert, shut up and get to work.

 

[Austin]

Koert: I am at work finishing up a large estate appaisal and just finished the third one today thinks to regression methods. Your above post is so loaded with questions that it would take a book to respond to all of them in detail. I will hit some tonight when I have time. I don't know about youse guys, but I am busier than I have ever been. The phone is not plugged in and they are walking in off the streets and won't take NO for an answer. I went to lunch and a buddy was begging me to go back into the real estate auction business and sell his farms. I got back to the office and a buddy I use to work with wants to put his license with me and get back into the brokerage business. Then I come on line to see if I have been slandered and bingo! The problem with being "right" is that there is always a gang out to get you. Later dudes.

 

[jtrotta]

George Hatch;

After reading your $.02 / $.02 - with regards to "Functional or External" obsolescence, I have a question??

Have we all established that either really fit the scenerio, perhaps there is no "Functional" or "External" - it would appear from the rebuttle's here, the possibility exists that neither are applicable.

We are on the approach of the Great Divide 8)

 

[KD247]

Austin,

I know what you mean. This was supposed to be a fairly relaxed week, but the fax machine has been barking out orders all morning.

Hey, thanks for having a conflicting and interesting viewpoint. Keeps me on my toes.

Koert

 

[Austin]

Koert: I will tackle your shotgun post from above one pellet at the time. First, the meaning in the definition of market value of “most probable price.” Whether you realize it or not, everything you do and every decision you make is based on statistical probability. When you get into your car, you could get killed in a wreck; when you get into an airplane you could crash, etc. Then why would a sane person drive or fly? The answer is that although these things can happen, the probability is so low that you are willing to take the chance. It is part of the human logical system of decision-making. You get out of the bed in the morning based on a statistical analysis of today’s events even though you don’t realize it.
Now, why do I say that “most probable” is a statistical definition? Because when we deal with random variables in our decision making process we never come up with a point estimate, but instead, we always come up with a range of values. For example: If I gave a set a of house plans to 20 appraisers from this board and asked them to use their cost manual to estimate the cost of the same new dwelling, I can guarantee you that we would come up with 20 different cost estimates. Your Marshall & Swift cost manual reports about 6 cost qualities of construction classifications, each of which is the center point of a range of costs from the market. If the dwelling I gave out to the appraisers to estimate was a 1,500 square foot rancher with full basement and carport, the reported cost from the appraisers would probably range from lets say $90,000 to $110,000. Why? Because each appraiser would make a different determination of quality of construction and second, the cost manual is based on statistical cost data. If we have a range of random variables ranging from $90 to $110,000, then what is the correct answer?
You have to look at it from a statistical perspective. When you gather data on any random variable it will always fall within a range. The more samples you use the closer the graph of the results will look like a bell curve. The data is trying to show you the true value of the variable from the population by statistical inference. If we have a range of variables as described above, then how do we know which is the correct answer? The answer is the probability density function as expressed by the bell curve. The most probable price is the one with the highest probability, which is the one under the probability density function or bell curve. That in my opinion is why the definition of market value specifically says “most probable price.” If we had the freedom to choose the price in the above example we would come up with 20 different price estimates. This is the only way to describe a range of random variables. The more samples we get, the closer we get to the correct answer.
When I first started appraising, the definition of market value used the phrase “highest price in terms of money.” What does that mean? Using the above example, it could only mean the highest price that one of the appraisers came up with in the cost approach ,I guess. Another definition we frequently hear is “fast sale price.” What does that mean? Without a probability density function as a reference, none of the three definitions just stated has any point of reference. When you say, low, most probable, and high-in relation to what? It has to be in relation to a probability density function to have meaning.
I have had this discussion many times before and I typically hear, “most probable price means the price that the appraiser feels is right.” In my above example, we came up with 20 different feelings. The logical way to deal with this situation is to view the 20 different feelings as each being a random variable under the bell curve and the most probable price is the one at the high point under the curve, that being $100,000 in my example. Then too, given the apparent intent of USPAP with all of the requirements, does it make sense to use all of those sophisticated appraisal methods and then give the appraiser the freedom to throw it all out the window and report a price based on how he/she “feels.” I think not. More later:

 

[Anonymous]

quote<< USPAP does not say anything at all about bracketting properties>> . Amen Brother.
As for being outside brackets, forecasting beyond or below any given indication is not only possible, legal, and sometimes sensible, in some industries, it is all you have. i.e.- appraising oil and gas properties REQUIRES a forecast of market prices for oil and gas. There is no way to "bracket" future prices upon which you base such appraisal. Does that make the appraisal impossible? indefensible? NO. USPAP & FIRREA DOES require you have knowledge of the future but does expect you to determine that the past is a predictor of the future. Even in the resort market to the east around Beaver Lake, once a mansion was built on 100-200 acres of land on shoreline. 28 bedroom, a tunnel between the servants quarters and the main house, etc. Man who brought Volkswagen to America built it about 35 years ago. It has set vacant for 10 yr and finally sold for several million after being offered, then auctioned, fought over, withdrawn, etc. Sold for 1/2 the assessor appraised value, mere 20% or so of estate appraisal did 5-7 years ago. STILL sold for far more than any house around the lake. Marketing time? 3-5 years maybe if you only count the time it was really on the market. Yes, boys a property will sell outside the brackets so it d--- sure can be appraised outside them.

As for some one's stated concern WHY someone would build or why a bank would let them. Let me state the not so obvious then the obvious. The land value "as if vacant and available for its highest use" exceeds the amount of money to be borrowed. So the bank does not give a hooey and a half-hitch about what the buildings cost. They know they can get their money back in a worst case scenario. But the banker and builder both share is a lack of appreciation for a basic concept we appraisers HAD TO BE TAUGHT OURSELVES. Namely, COST and VALUE are not one and the same. Both know what the land COST. Both know what the buildings will COST. NEITHER understand why 1 + 2 = something less than 3. (or possibly the banker does understand but was being differential to the customer.) The builder has not been hung with one of these in the market he is building in (a mere 25 mi. away) but he will be hung with this monster should he decide to sell it. He likely will not. Why did he chose here? His church is next door, his kids school is 3 mi. away (private Seventh day Adventist school). THIS IS WHERE HE WANTS TO LIVE. The notion that we or the bank or even common sense would prevent someone from building what ever they want anywhere they want to is nonsense. Otherwise NOBODY in their right mind would build a Dome house, Plastic House, one-of-a-kind home, etc. etc. They do it all the time. And if we did not say no, and the bankers say no to all the people who don't have 50% LTV money to put down, there would be a ton more of them built. And if we are DOING OUR JOB we will identify and quantify to the best of our ability, within our stated scope of work, based upon "the quantity and quality of data AVAILABLE" [direct USPAP quote] estimate a value within a stated marketing time consistent with the market.
And one last thing. I alluded above to a near decade exposure of a mansion above. In the real world where people borrow money, a marketing time rarely exceeds 3 years before the bank either forces sale or gets the property back. It is then liquidated. This liquidated price is usually for the mortgage or maybe mortgage plus legal expense at the best. Which of course means that it still does not make good comparable data.
Terrel

 

[Anonymous]

Austin said <<[how can] you achieve credible results by using a technique (extrapolation) that is known not to produce credible results? >> and said this is the gist of Std 1-1 abc. Wow. I am sorry, but I cannot see extrapolation written even in the margins or definitions.

You may argue a bell curve for sales, but when you apply sales cumulatively the curve is S shaped not bell shaped.