Broker Price Opinions in Nevada

[George Hatch]

Bert,

I'm sorry, are we still talking about AVMs? I thought we were talking about appraisals. I'd imagine that the reason a lot of appraisers use "Accurate" in their company titles is because it comes up in the Yellow pages before "Adequate" or "Reasonable". The fact that we have a lot of appraisers who don't give the subject much thought doesn't speak well for us as a group. I don't think it's anything to laugh about.

In the current USPAP parlance, we do opinions, not estimates; and our standard is reasonable analyses, opinions and conclusions, not accurate. An AO commenting on checking the accuracy of an AVM doesn't trump the SOW Rule or SR1 or SR3-1.

That may change in the future, but for now the credibility of a workproduct is always measured within the context of intended use. As far as I can tell, some of the resistance to USPAP from appraisers stems from their overuse of absolutes. There's often a backlash.

Establishing credibility starts with establishing reasonable (there's that word again) expectations. Appraisers get into trouble when they start promising more than they can deliver. That's the problem Zaio ran into - they have been making claims they can't back up.


It may be the right thing to do to force lenders to use appraisers for ALL of their valuations, and for appraisers to be limited to only one or two options with their workproducts. However I'd say that so far we've been struggling (as a group) to do the right thing just with the work that we already get. It's not like we have the moral high ground to support a drive to cut off the competition.

No matter what the lenders use and no matter what appraisers are allowed to do, there will always be compromises between purity vs. expediency. The balance between the two has always shifted with the winds of economic trends. Right now it looks like those winds have shifted in our favor due to the failure of the mark-to-model orientation these lenders have been using. That won't last forever, though, and we won't be able to hold onto any gains we've made as a result of chasing trends. I really do think we need to take the broader view and pay more than lip service to "meaningful to the intended users".

The lenders already consider us to be a tax on their commerce, inflicted on them by the government. In their eyes we are a special interest group. I think it would be to our benefit if we could try to avoid reinforcing that perception.

But all of that is just my opinion - you guys are obviously entitled to your own opinions. There's nothing wrong with appraisers acting within their own self-interests so long as they recognize that it's in their own self interests to adapt a bit to changes in what these intended users consider to be meaningful. IMO.

 

[xmtraterbcra]

... our standard is reasonable ....

George,

Your comments are good and I can't say I really disagree that much.

However, I would like to make the comment that I CAN give an operational definition of "appraisal accuracy". It's similar to the task that physicists have in trying to identify sub-atomic particles. Although they can never "see" them, they can set up tests to verify their existence. True "Market Value" is kind of like a sub-atomic particle. It does exist - but it's hard to identify directly. We can get at it though, through a somewhat lengthy and indirect process. - An Appraisal Registry would certainly make this a real possibility.

With regards to USPAP "reasonableness" - it really does not have a definition, certainly not an "operational definition". It's a mish mash conglomeration of a number of things - and I dare say difficult to enforce. Therefore, USPAP should extend itself to deal with concepts of accuracy, probability-value curves and the like. New territory that needs to be explored - and for which future tools coming on the market (sooner rather than later, I hope) should help.

Bert Craytor, SRA

 

[George Hatch]

Bert,

I'm a USPAP Instructor - all I've been doing here is promoting the party line with respect to the applications thereof. When the party line changes then I'll change with them.

From my perspective, BPOs-as-evaluations don't create a hazard because realty agents are doing them; they're a hazard because these individuals who are acting as appraisers (and thus fit our definition of an appraiser) are not adhering to the professional standards for appraisers. We define the applicability of those standards by the actions of the individuals and the expectations of the users; not the labels, nomenclature, titles and licensing statuses involved.

--------------
As for "accuracy"......

I don't disagree with you that comparisons between value conclusions and resale value can be used to build confidence scores, but doing that does require the use of certain assumptions, interpolations and extrapolations. The SFR market is more homogenous than any other market segment, but there's still an awful lot of covariance going on.

This is where the statistical analyses elements would come in. FWIW, I don't think RA is going away and I don't think the appraisal profession has yet reached it's peak in terms of progression. In my mind this reinforces the utility of keeping an open mind and taking a broad view of the underlying concepts and principals involved in our professional standards.

 

[xmtraterbcra]

As for "accuracy"......

I don't disagree with you that comparisons between value conclusions and resale value can be used to build confidence scores, but doing that does require the use of certain assumptions, interpolations and extrapolations.

George,

Actually that is not such an issue because I am suggesting that we AVERAGE the individual inaccuracies of all appraisals. IF the appraiser is unbiased in his work, then the anomalies related to each appraisal will average out.

That is to say, assume we have an Appraisal Registry that requires every appraiser to register the appraisal, with his "opinion" of market value. We also have access to propety records, via say NDCData. All the data is copied and stored in a SQL database. We crossreference all appraisals per each appraiser against property records sales within N years following each appraisal. Maybe we only get 10-20 sales that the appraiser had done appraisals for 6+ months prior. We time adjust the sales (ignore other market conditions). Yes, the time adjustments are not perfect. But, they are largely indiscriminant. If some are too high, then there are probably an equal number to low. The law of averages. So, the average of the percentage inaccuracies is probably a fairly accurate measure of the appraisers overall percentage inaccuracy. If we do this for all appraisers who work in the same market area and rank the results, we get a ranking that can be used as a measure of the appraisers accuracy: For example, 15% of the appraisers in the given market area are more accurate and 85% less accurate. The appraiser accuracy score: 85%. We could even convert this to an Appraiser IQ!! And, .... YOU CAN'T CHEAT HERE. Enforcement then could focus on those appraisers with low scores - and essentially work from the bottom up.

To me this make sense. A lot of sense. What are typical appraisals for? They are used to make decisions about collateral. Quite frankly investors are more interested in what property is going to be worth a number of years out from the date of loan - because that is when defaults are most likely to occur. The market can change in that time. But their people, Collateral Managers, know how to estimate the risks that the market is going to go up or down. They review their portfolios periodically against changing market conditions and assess risk. The appraised value (read opinion of value) is only an input into their system. It is ground zero. But it is invariably adjusted for time and possibly other market conditions before being used in the risk assessment analysis. This IS exactly what we should be measuring.

"Reasonableness" is just a play toy that does nothing more than create a nice ambience for appraisers. It is not a measure of anything. If USPAP can't even define "reasonable" or "reasonableness," what more can one say?


BTW, I suggest appraisers carry out this kind of analysis on their own appraisals. The results might be interesting for them.

Bert Craytor, SRA

 

[Couch Potato]

It is unreasonable to expect USPAP to define every word it uses.

 

[xmtraterbcra]

It is unreasonable to expect USPAP to define every word it uses.

Yes, of couse. But USPAP should be defining core terms that are heavily used. The point is that "reasonable" is a term that has to be more specifically defined in each context it is used. In nearly all cases it is not. I can assure you that one appraiser's understanding of what "reasonable" means in a given context, is generally not the same as another's. The result is that its meaning in any particular context is at the mercy of the reviewer(s). USPAP is supposed to be a STANDARD. "Reasonable" is not a standard by any measure.

So, we hear from USPAP instructors: "We cannot use the term appraisal accuracy, because that would imply a level of accuracy that is impossible to attain." You can always provide a definition of accuracy and put it in your reports that very clearly states that accuracy is such and such. You can provide operational definitions that fit the bill precisely. Scientists do this all the time. No problem. It's nonsense to say otherwise.

To be fair an honest, USPAP designers have gone so far ... but not quite far enough.

Let me improve upon USPAP:

REASONABLE (Definition): The specific meaning of REASONABLE depends on the context in which it is used. For each context in which the term REASONABLE is used, there should be an OPERATIONAL DEFINITION, i.e. a definition that is based upon a prescribed set of operations that will determine if whatever is described as REASONABLE is in fact reasonable.

The key here is the concept of OPERATIONAL DEFINITION as employed by physicists and others. You can create a hierarchy of operational definitions, e.g. a REASONABLE opinion of market value requires X, Y & Z. X requires REASONABLE X1, X2 and X3. Etc..

Naturally the objection here is that USPAP would expand in size by 4 or 5 fold. I would contend that we don't need operational definitions for everything, just for some core terms and contexts, as examples that show how to construct them for the innumerable contexts we find in the market. USPAP should state that where it does not provide OPERATIONAL DEFINITIONS, the appraiser should construct those to reflect what he means. ... This would be part of the SOW.

Bert Craytor, SRA

 

[Fred]

For example, 15% of the appraisers in the given market area are more accurate and 85% less accurate. The appraiser accuracy score: 85%.

Isn't that more or less what ERC companies do?

The problem with what you are proposing is that it is as ponderous a task as the one Zillow undetakes. You would need a "statistical" measure, and I don't mean raw percentages. Hitting prices with a certain degree of accuracy might be lousy work on one assignment and brilliant work on another with both properties located within what might appear to be the same "neighborhood."

 

[Fred]

It is unreasonable to expect USPAP to define every word it uses.

It would become more reasonable if USPAP used less words.

Just off the top of the old cabesa, since 1991, synomyms for whether some analysis (or approach) should be developed or relied on by a client.
Suitable -which only appeared in the 2006 USPAP
Applicable - always appeared in the document, defined indirectly 1999-2005, not clear if the erstwhile defintition is still "applicable" to the current usage.
Necessary- always in the document, defined 1999-2005
Credible - always in the document, defined since 2005
Relevant - not defined, not sure of the legacy
Meaningful - always in the Preamble
Reliable - in SMT-7, 7/93 - 12/03, not defined
Appropriate - in the SOWR, alway part of SR 1-4.
Accurate - in the Competency Rule, not sure of the legacy, (not development, but part of Std 2)

Also, Std-3 requires reviewers to check for reasonableness and completeness, and some other things that don't appear to be required of the original work by Std 1 or the SOWR.

So, to me, it is well within reach to have all the key terms that establish connections between one concept and another defined, even it if means simply throwing out the ones that don't have defintiions.

As to accuracy. I have no problem applying "accuracy" to "most probable price" assuming prudent and knowledgeable parties. George and other long-time forumites might remember that I have long advocated using something akin to "SWAG value" on some assignments. I continue to feel there is a disconnect between using a value defintion that refers to prudent and knowledge parties in assignments where the conditions effectively "scope" away the diligence necessary to become prudent and knowledgeable. For example, instead of Zillow using the ASB's ridiculous opinion (#18) as cover, they should meet standards by stating the type of value opined - WAG value - a cut below SWAG.

 

[xmtraterbcra]

Isn't that more or less what ERC companies do?

The problem with what you are proposing is that it is as ponderous a task as the one Zillow undetakes. You would need a "statistical" measure, and I don't mean raw percentages. Hitting prices with a certain degree of accuracy might be lousy work on one assignment and brilliant work on another with both properties located within what might appear to be the same "neighborhood."

I don't know. But, I would suppose they are using prospective value (based on market time) as the basis for their measurements. That's a lot easier than having to wait a couple of years after an appraisal for a sale to possibly occur.

Bert Craytor, SRA

 

[woodyva]

I would be satisfied if USPAP stated simply that it is not a guide to protect the public trust, but is set forth to protect the lenders and lender lackey's that maintain it at the AF.

I know it sets out to be public trust specific, but it is more or less in place due to a lender crisis and the reaction to it. Their inability to keep up with technology make sit suspect. A good friend of mine who is a respected appraiser as well was having this conversation with me the other day. If they did keep up with the tech, they would most likely do what the lenders wanted, not what wa sin the best interest of general public.