The Lie

[DiverMike]

Hi Mike,

I believe in some law, oh sometime between 2011 and 2015, they wrote some rule and definition that said opinions can only be formed by humans. Hence, AVMs are not appraisals as the computer can not perform an opinion, it can only provide a range, and that would be a ditto for a computer review.

You wouldn't think they left their bases uncovered since they have been working on this since 2005.

.

Marion, A rating founded on an automated SYSTEM that is a reflection of the personal opinions and interpretive view of its software designers is ultimately no more than the result of THEIR opinions when they designed it. A "rating on the quality" of an appraisers work IS still an opinion or STATEMENT whether done by a human being or through a mechanized or computerized program. It is a defined "process" which IS covered. AVMs performed and communicated by appraisers ARE appraisals. Respectfully you need to recheck your definition of what an appraisal or an appraisal review is. Now a few years ago they were estimates of value, but now they are opinions again due to ever changing fundamental TAF appraisal "Principles". In California if I am asked for an AVM (as an appraiser), I can run it off; print it out but offer NO OPINION or even explanatory comments on it. Otherwise It IS subject to state appraisal laws. I cannot even cite the visibly obvious range of value.

You are quibbling. The real issue is that the system is flawed whether you feel review can only be opinion or whether as a practical matter it can derive from automation. The "non" opinion "rating" is initially based on an AVM which when applied or contrasted to an appraisal is MEANINGLESS! THAT is why we HAVE review standards in the first place! Otherwise replace ALL of us and CU with ZILLOW! Then they back up the 'rating or automated conclusion" with purported peer results from the same CLAIMED geographic area EXCEPT that they do not acknowledge the peer database was created back when even FNMA admits that MANY if not MOST appraisers were appraising to the guidelines rather than to the market. Now they are comparing those subjective RELATIVE ratings (good, average, fair, fair to average, etc.) to ABSOLUTE ratings using a completely different system! How can they POSIBLY claim "peer adjustments" in good faith or honesty? They also fail to effectively identify that the default CU is based on census tract rather than the SAME neighborhood or competitive market area defined by the appraiser.

I wouldn't think they'd screw up a software program they worked on since '05? Marion When I worked at IRS in '09 and '10 when they got authorization to move up to Office 2003 it was already an already obsolete program. Another infamous computer fiasco had to do with the Affordable Care Act, remember? Money and time are no guarantee of quality or competency when it comes to big bureaucracies. WHY would I believe an appraisal "rating" software program worked on by individuals that did not have ONE ACTIVE appraiser license among them could be flawed? Seriously?

FNMA has twisted itself into a pretzel to pretend this system is a valid measure of ANYTHING in untrained hands. They've failed miserably.

 

[DiverMike]

Hi Mike,

I believe in some law, oh sometime between 2011 and 2015, they wrote some rule and definition that said opinions can only be formed by humans. Hence, AVMs are not appraisals as the computer can not perform an opinion, it can only provide a range, and that would be a ditto for a computer review.

You wouldn't think they left their bases uncovered since they have been working on this since 2005.

.

Marion, A rating founded on an automated SYSTEM that is a reflection of the personal opinions and interpretive view of its software designers is ultimately no more than the result of THEIR opinions when they designed it. A "rating on the quality" of an appraisers work IS still an opinion or STATEMENT whether done by a human being or through a mechanized or computerized program. It is a defined "process" which IS covered. AVMs performed and communicated by appraisers ARE appraisals. Respectfully you need to recheck your definition of what an appraisal or an appraisal review is. Now a few years ago they were estimates of value, but now they are opinions again due to ever changing fundamental TAF appraisal "Principles". In California if I am asked for an AVM (as an appraiser), I can run it off; print it out but offer NO OPINION or even explanatory comments on it. Otherwise It IS subject to state appraisal laws. I cannot even cite the visibly obvious range of value.

You are quibbling. The real issue is that the system is flawed whether you feel review can only be opinion or whether as a practical matter it can derive from automation. The "non" opinion "rating" is initially based on an AVM which when applied or contrasted to an appraisal is MEANINGLESS! THAT is why we HAVE review standards in the first place! Otherwise replace ALL of us and CU with ZILLOW! Then they back up the 'rating or automated conclusion" with purported peer results from the same geographic area EXCEPT that they do not acknowledge the peer database was created back when even FNMA admits that MANY if not MOST appraisers were appraising to the guidelines rather than to the market. Now they are comparing those subjective RELATIVE ratings (good, average, fair, fair to average, etc.) to ABSOLUTE ratings using a completely different system! How can they POSIBLY claim "peer adjustments" in good faith or honesty?

I wouldn't think they'd screw up a software program they worked on since '05? Marion When I worked at IRS in '09 and '10 when they got authorization to move up to Office 2003 it was already an already obsolete program. Another infamous computer fiasco had to do with the Affordable Care Act, remember? Money and time are no guarantee of quality or competency when it comes to big bureaucracies. WHY would I believe an appraisal "rating" software program worked on by individuals that did not have ONE ACTIVE appraiser license among them could be flawed? Seriously?

FNMA has twisted itself into a pretzel to pretend this system is a valid measure of ANYTHING in untrained hands. They've failed miserably.

 

[DiverMike]

Potshots?

Fannie Mae is using THEIR proprietary software and data to QC appraisal reports in real time versus finding out later that the appraisal was crap and then trying to get their contracted lender to buy back the loan or find out they've got tens of thousands of bad properties in their portfolios. This has nothing to do with the term Appraisal Review and Appraisers.

CANative, IF that were purely the case, then why would appraisers receiving high scores be subject to being placed on a watch list even after the specific appraisal(s) that are rated "high" were clarified in writing? The CU score does not change, it merely is over ridden in such cases. It is STILL subject to post purchase audit adjustment (buy back). FNMA CU course data says FNMA will still buy the loan with even a 4 or 5 rating; but as a practical matter lenders are killing deals with a CU higher than 3.5. Apparently they can't get a "clean" SSR above that.

IF the CU system were ONLY used by and available to trained appraisers that made the over ride determination OR recommended referral for field review, I'd have NO problem with CU. If it were available to appraisers BEFORE submission to FNMA (or client which causes an immediate upload and scoring) I'd have no problem with it. One of its tools (graphic map alternate comps function) would possibly prevent mal inclined appraiser from ever again comp shopping for the few that can 'make the deal happen' but at same time would let them KNOW what's out there that no one would NORMALLY think to explain, but when CU is applied SHOULD be explained...like those up to 20 alternate comps! I'd NEVER explain non-use of 20 comps dating back three years!

 

[CANative]

How can you be so smart and yet get this so wrong? It's a black box that Fannie Mae uses for it's own internal decision making. It's output is not an Appraisal Review. It's an "appraisal review."

then why would appraisers receiving high scores

It doesn't score appraisers. It scores their work work product (I suppose it has THAT in common with what an upper case Appraisal Review is.) Any intended user of an Appraisal Report can have requirements and assignment conditions that exceed USPAP and that includes Fannie Mae. I think a lot of appraisers don't understand this. The black box is designed to look for this and that is because they have an obligation to their investors and regulatory requirements. These requirements are published publicly and are easily available for anyone to see. It would be irresponsible of Fannie Mae to not have some sort of system to ensure compliance. But because now they have to vet the appraisal reports in real time while lenders and borrowers are waiting for approval there has to be automation and that's what CU does.

Appraisers should NOT have access to raw output. It might influence their decisions. They would appraise defensively and select "clean comps" that won't cause problems. Just like the 15/25 adjustment dynamic.

I don't think Fannie Mae is behind the problems between lenders (who use the system), AMCs and appraisers. I think it is more a case of lender underwriting not using the system in accordance with their agreements with Fannie Mae and Fannie Mae's rules for using it.

If I were more cynical and rude I'd say your article is pandering to the fears and mistrust of "the government" that many appraisers feel. I think that fear and mistrust is irrational.

 

[Meandering]

Marion, A rating founded on an automated SYSTEM that is a reflection of the personal opinions and interpretive view of its software designers is ultimately no more than the result of THEIR opinions when they designed it. A "rating on the quality" of an appraisers work IS still an opinion or STATEMENT whether done by a human being or through a mechanized or computerized program. It is a defined "process" which IS covered. AVMs performed and communicated by appraisers ARE appraisals. Respectfully you need to recheck your definition of what an appraisal or an appraisal review is. Now a few years ago they were estimates of value, but now they are opinions again due to ever changing fundamental TAF appraisal "Principles". In California if I am asked for an AVM (as an appraiser), I can run it off; print it out but offer NO OPINION or even explanatory comments on it. Otherwise It IS subject to state appraisal laws. I cannot even cite the visibly obvious range of value.

You are quibbling. The real issue is that the system is flawed whether you feel review can only be opinion or whether as a practical matter it can derive from automation. The "non" opinion "rating" is initially based on an AVM which when applied or contrasted to an appraisal is MEANINGLESS! THAT is why we HAVE review standards in the first place! Otherwise replace ALL of us and CU with ZILLOW! Then they back up the 'rating or automated conclusion" with purported peer results from the same geographic area EXCEPT that they do not acknowledge the peer database was created back when even FNMA admits that MANY if not MOST appraisers were appraising to the guidelines rather than to the market. Now they are comparing those subjective RELATIVE ratings (good, average, fair, fair to average, etc.) to ABSOLUTE ratings using a completely different system! How can they POSIBLY claim "peer adjustments" in good faith or honesty?

I wouldn't think they'd screw up a software program they worked on since '05? Marion When I worked at IRS in '09 and '10 when they got authorization to move up to Office 2003 it was already an already obsolete program. Another infamous computer fiasco had to do with the Affordable Care Act, remember? Money and time are no guarantee of quality or competency when it comes to big bureaucracies. WHY would I believe an appraisal "rating" software program worked on by individuals that did not have ONE ACTIVE appraiser license among them could be flawed? Seriously?

FNMA has twisted itself into a pretzel to pretend this system is a valid measure of ANYTHING in untrained hands. They've failed miserably.

DON'T ARGUE WITH ME ABOUT IT. IT HAS ALREADY BEEN ASKED AND ANSWERED AND WE LOST THE ARGUMENT 5+ YEARS AGO.

IAEG 2010.​

Reviewing Appraisals and Evaluations. This section in the Proposal and the Guidelines provides the Agencies' expectations for an institution to establish an effective, risk-focused process for reviewing appraisals and evaluations prior to a final credit decision. In the Proposal, the Agencies specifically requested comment on the Agencies' expectations for reviewing appraisals and evaluations. In particular, the Agencies sought comment in the Proposal on whether the use of automated tools or sampling methods for reviewing appraisals or evaluations supporting lower risk residential mortgages are appropriate for other low risk mortgage transactions. The Agencies also requested comment on whether appropriate constraints can be placed on the use of these tools and methods to ensure the overall integrity of the institution's appraisal review process for other low risk mortgage transactions. Commenters requested further clarification on the process for institutions to obtain approval to use automated tools and sampling methods in the review process. The Proposal noted that each Agency would address the approval process through established processes for communicating with its regulated institutions. ...

You can read it the whole thing, too long to post here.

If you want it reconsidered (fat chance) you have to take it up with the Inter Agencies.


.

 

[Terrel L. Shields]

Otherwise replace ALL of us and CU with ZILLOW!

I preferred to replace them with real clients.

FNMA has twisted itself into a pretzel to pretend this system is a valid measure of ANYTHING in untrained hands. They've failed miserably

How do you admit failure for your own creation? It's their internal "problem"...not mine.

If I were more cynical and rude I'd say your article is pandering to the fears and mistrust of "the government" that many appraisers feel. I think that fear and mistrust is irrational.

Yes, Can, I trust the government. Re: my signature from a 2012 campaign speech by the POTUS. Fat lot he knew.

 

[CANative]

> sigh

 

[Deleted member 80407]

George Orwell 1984. The government makes the law, the government follows the laws it wants to.

I agree. Want to read a good book? Read " The Creature of Jekyll Island" I promise you, it will freak you out.

 

[Randolph Kinney]

I agree. Want to read a good book? Read " The Creature of Jekyll Island" I promise you, it will freak you out.

I have. The backroom deal to create the federal reserve. Just shows you how corrupt the system is.

 

[timd354]

It would be "illegal" (in my opinion) because it IS an appraisal review and one that is being performed by unlicensed personnel via "a system or process" that does not meet ANY USPAP standards for real estate appraisal review, by a federally regulated institution; GSE. It is "illegal" because it is misleading and fails to communicate the results of an appraisal review in a meaningful or honest way. It is fraudulent (again, in my opinion) because it is used in part to assure purchasers of bundled loans or securities that the loans they are buying have been subjected to a meaningful or relevant valid "collateral risk rating system" that is fundamentally flawed and incapable of accurately delivering what is promised.

Thanks for a good laugh today...you obviously are unable to distinguish between a standard 3 appraisal review and a QA appraisal review.