The Race To Automate Appraisal Process

[George Hatch]

Nine out of ten times, when it favors their agenda, brokers will put forth a price per square foot analysis in the ROV.

What you're specifically referring to is the price/sf GLA. There's also a price/sf lot area for each one of those sales, too. And a price/room. These are simply units of comparison, just the same as the sales price itself is a unit of comparison when used that way in the Fannie grids. No more, no less.

And how do the brokers use it? If they do ANY analysis they rank the subject in the range. Superior to these two and maybe a little inferior to this one. And NOW is a great time to buy because I've had multiple all-cash offers for more, so this is a great deal.

 

[Carnivore]

BTW, most buyers and sellers in the SFR markets who are developing their own opinions of value (as opposed to taking their broker's word for it) employ a qualitative analysis and rank the subject among their comps. Virtually all brokers on the residential side use a qualitative analysis in their valuations. That's the methodology that is actually used in the market by its participants.

That is the variable that can't seem to be captured. Google Analytics uses our data input to try and foresee what advertisers are sent your way while you cruise the internet. With me they often get it wrong....but continue to send me advertisers long after I have either purchased or changed my mind.

I can make one small improvement to the FNMA series that would in one way prove very helpful to AVM companies. My suggestion to the form would fill some gaps or holes in the data. Right now we explain something in writing. A simple added checkbox would add a lot for AVM's. We kind of do it now, but its real time, not historical.

I am not againt's Regression at all.It can prove to be a very valuable tool, but I think I understand its limitations. I don't fully understand the process. I do know as George pointed out the process can be different from one model to the next.

 

[J Grant]

The problem is the cheap fast/model is infecting everything and appraisers who choose to use RA /auto populate without understanding the results will be rewarded by given assignments, with RA nothing more than a tool to get them done fast and cheap.

An RA or other auto population data dump allows an appraiser to get fast results with a pseudo impressive "proof" of "support" for adjustments. Esp applicable for hybrid work which is fast turn time and lower pay. It won't matter to appraisers that they don't understand the RA or USPAP section about understanding it.

The economic pressures on lending work with AMC's or other low pay venues means it is really the appearance of meeting min standards that counts...since min standards means credible , how would an appraiser sitting at a desk who rarely sees properties and use data dumped RA or auto populated software know what is credible, and if they want to know, they are not being paid to spend the time to find out. It is possible they might face consequences at some point if results are faulty, but that is a distant future risk they are willing to take, or they are deluded there is no risk since supposedly the RA provides bullet proof adjustment (or whatever nonsense the RA seller promised).

 

[J Grant]

Correct in that "traditional" methods is modeled on buyer/seller behaviour rather than just data. But the problem is the cheap fast/model is infecting everything and appraisers who choose to use RA /auto populate without understanding the results will be rewarded by given assignments, RA is a tool to get work done fast nothing more. Sometimes it might get A) great results, other times B) mediocre results, other times C) bad results, the problem being it is a crap shoot when A. B or C happens. .

An RA or other auto population data dump allows an appraiser to get fast results with a pseudo impressive "proof" of "support" for adjustments. Esp applicable for hybrid work which is fast turn time... It won't matter to appraisers that they don't understand the RA or that there is USPAP section about it,

The economic pressures on lending work with AMC's or other low pay venues means it is becoming the appearance of meeting min standards that counts...since min standards means credible , how would an appraiser sitting at a desk who rarely sees properties and use data dumped RA or auto populated software know what is credible? And if they want to know, they are not being paid to spend the time to find out on any one assignment. It is possible they might face consequences at some point if results are faulty, but that is a distant future risk they are willing to take, or they are deluded there is no risk since supposedly the RA provides bullet proof adjustment (or whatever nonsense the RA seller promised).

 

[George Hatch]

I don't know that cheap/fast is necessarily the driver here. I think that much of it is a backlash against lazy appraisal practice.

Straight "rank the subject in the range" qualitative analysis works, but only when someone actually performs that analysis.
"Appraiser's judgement" works, but only when someone actually performs that analysis
"Matched paired analysis" works, but only when someone actually performs that analysis.
"Group analysis" works, but only when someone actually performs that analysis.
"Sensitivity analysis" works, but only when someone actually performs that analysis

Etc, etc. They all work to one degree or another - but only when someone actually does it. No analytical method works when they are merely referenced in the report but aren't otherwise performed. To the extent that so many of us have not been doing what we say we do, that's why the charts and graphs crowd has been able to sell the chrome plated spinners with the 22"s as a necessary component of the Yugo we are selling to these lenders.

We did this to ourselves.

 

[J Grant]

I don't know that cheap/fast is necessarily the driver here. I think that much of it is a backlash against lazy appraisal practice.

Straight "rank the subject in the range" qualitative analysis works, but only when someone actually performs that analysis.
"Appraiser's judgement" works, but only when someone actually performs that analysis
"Matched paired analysis" works, but only when someone actually performs that analysis.
"Group analysis" works, but only when someone actually performs that analysis.
"Sensitivity analysis" works, but only when someone actually performs that analysis

Etc, etc. They all work to one degree or another - but only when someone actually does it. No analytical method works when they are merely referenced in the report but aren't otherwise performed. To the extent that so many of us have not been doing what we say we do, that's why the charts and graphs crowd has been able to sell the chrome plated spinners with the 22"s as a necessary component of the Yugo we are selling to these lenders.

We did this to ourselves.

Oh please we did not do this to ourselves, and most appraisers do what they said re matched pairs or other analyses. Appraisers are far from lazy, the amount of work that goes into even a mediocre URAR is time consuming -and underpaid, relative to the $ other make in a RE transaction.

How did we "do this to ourselves , when big data and auto software to perform RA did not exist a mere decade ago, as it has in recent past years when it became mass market available and affordable? WE did that? Give me a break.

Add in that post HVCC appraisals became hijacked from a collateral service to be an income stream for AMC;s . The AMC's will apply the same profitability criteria for which product to use- if they can make more money pushing alternate products or weakened appraisals like hybrids to clients, then that is what they will promote. Now that for profit parties have become the major "stakeholders", it will be a profit driven decision that determine what is acceptable as min standards, and will drive decisions around evaluation or appraisal products are chosen, and then which appraisers or analysts or inspectors are hired.

 

[Basically Rectangular]

The appraiser really should understand the AVM and how it derives its adjustments. There should be rules for the client (AMC) that is providing the proprietary AVM software to include training or at least a manual that appraisers can reference when using the AVM. Otherwise we are just trusting the adjustments to be correct.

To me the AVM is a form of an approach to value, such as the "modified sales comparison approach". Is there USPAP guidance on this? I need read the latest....

While the fees for these are low right now, I can see them increasing in the future. My logic is that there is a considerable market and too many parties profiting off of appraisals to simply decrease all of the fees. Are the AMC's going to be swept aside when lawmakers flip the switch and standardize these products? I'm speculating here, as I do not have access to lender balance sheets or inside info, but I do not think that the lenders will necessarily want lower fees since many may have an interest in AMC business, either directly or indirectly. They are too smart to not be making a risk-free profit off of the appraisal process.

 

[George Hatch]

Oh please we did not do this to ourselves, and most appraisers do what they said re matched pairs or other analyses. Appraisers are far from lazy, the amount of work that goes into even a mediocre URAR is time consuming -and underpaid, relative to the $ other make in a RE transaction.

Not every problem in our business is attributable to AMCs. I know you hate them, but there are other elements at play, too. One being that advances in the development and reporting always occur in the aftermath of a market bust as the lenders try to figure out how to avoid their previous problems.

 

[J Grant]

Not every problem in our business is attributable to AMCs. I know you hate them, but there are other elements at play, too. One being that advances in the development and reporting always occur in the aftermath of a market bust as the lenders try to figure out how to avoid their previous problems.

The "problem" in our business, whether it be AMC;s or software developers or emerging tech companies, be it AVM 's or cloud based solutions or hybrids or cheap fast appraisals, (whatever they are selling.) it boils down to appraising /evaluations, since for third parties have made them a revenue steam, the profession in part has morphed from providing a professional a service for borrowers/investors to providing a profit stream. for middlemen/re sellers.

For profit companies will promote and lobby for any product or fee structure that returns maximum to them and minimum to those they engage/ hire. Public trust is nothing more than a sound byte .Which puts ethical lenders and clients who use due diligence at a competitive disadvantage, same as appraisers who take time for research and credible results are at a competitive disadvantage, as for profit middlemen gain greater market share and voice in who gets hired or engaged.

 

[Eli]

The "problem" in our business, whether it be AMC;s or software developers or emerging tech companies, be it AVM 's or cloud based solutions or hybrids or cheap fast appraisals, (whatever they are selling.) it boils down to appraising and now evaluations, since third parties on large scale became involved, has morphed from a service for borrowers/investors to a profit stream. for middlemen/re sellers.

For profit companies will promote and lobby for any product or fee structure that returns maximum to them and minimum to those they engage/ hire. Public trust is nothing more than a sound byte .Which puts ethical lenders and clients who use due diligence at a competitive disadvantage, same as appraisers who take time for research and credible results are at a competitive disadvantage, as for profit middlemen gain greater market share and voice in who gets hired or engaged.

Lord, don’t get Mike Kennedy going on the bullseye. He will swear RA and AVM’s use that bullseye.

On the other hand, that’s where Crawford (and I know others here locally) are looking at opportunities with buyers, sellers, etc against these AVM’s and other valuation methods.

I believe it because I get more and more appraisals handed to me by buyers and sellers already. Recent appraisals at that. Sometimes agents hand them to me too.