CU Process Will Improve Appraisals? A Pipedream or Reality?
Fannie Mae Claims the CU Process Will “Improve Appraisals.” A Pipedream or Reality?
Appraisers,
A few days ago, I watched the FNMA eLearning webinar “Introduction to CU” video showing how their new Collateral Underwriter will be implemented, starting on Jan. 26, 2015, applying to 1004 and 1073 Reports (initially).
You really should watch this also! CU applies to FANNIE MAE only; FreddieMac, VA, FHA, USDA are not involved at this time.
Click this link, and you’ll find the link to the video ½ way down on the left.
I captured a few screen shots of their presentation to help explain what will happen, and have added comments below:
This is the basic process. Your appraisal (in .xml format) will be submitted by the AMC or the Lender to the UCDP portal. The UCDP will run a ‘rules set’ for UAD compliance, plus perform the new CU process to electronically review other aspects of your report. This will include generating a list of up to 20 additional properties considered to be ‘possibly comparable’ by the CU data base and FNMA models as shown below. The ‘user’ can then decide to send the report back to the appraiser, represented by the Dashed Red Lines (my addition to the screen shot), and ask for additional analysis and/or inclusion of one or more ‘comps’ into the report. Then your report has to be re-submitted back through the UCDP for final clearance.
Below is a sample of what the ‘user’ will receive from the CU electronic review – the list of appraiser reported comparables, and any additional ‘model-selected sales’ properties:
One aspect hidden in the CU process is FNMA’s property data base is tied to Census Tracts, and ‘value heat maps.’ So they can analyze property sales geographically very tightly in any area. While we don’t know yet for sure, the “perspective” I read between the lines is FNMA believes Census Tracts can be a defining aspect of comparable property selection. As I have previously written, when was the last time you performed comp selection research based around Census Tracts? Here’s a screen shot of Census Tracts in my area. Some are small and others are quite large. I can say that in all the years as an appraiser, I have NEVER considered comps based on Census Tracts:
Here’s what their ‘Census Tract Heat Map’ will look like (this is not in my area):
Also notice that the CU process will show the ‘user’ property value trends in the form of a graph based on the CU provided ‘model-selected sales.’
The most troubling aspects of this CU process are three:
- secrecy on the part of FNMA’s non-disclosure of (sometimes unreliable) market data until after the report is submitted to the UCDP;
- attempts to turn appraisal reports into entirely scientifically and mathematically supported discourses based on potentially questionable properties the experienced appraiser might consider inappropriate, or even appropriate, as to comparability, and
- insertion of the ‘user’ into the process to possibly manipulate the appraisal report, thereby removing mandated appraiser independence (in the Dodd-Frank law) from the entire process.
FNMA claims the CU process will “improve appraisals.” A pipedream, or reality? That remains to be seen, starting on Jan. 26, 2015.
Dave Towne on e-AppraisersDirectory.com
- New UAD Overhaul: What Appraisers Can Expect in 2025 & Beyond - September 19, 2024
- Cindy Chance Terminated - September 16, 2024
- Key Part of USPAP Not Available from TAF - July 19, 2024
With CU appraisers will be engaged in a game of high stakes poker with every appraisal they complete. Can you win a game against the government’s super computers? Sure you can. Can you win against the computers every time? Not a chance. Just remember each day that the bet you’ve placed on the table is your license and your livlihood.
Happy Gambling!
Freddie is implementing this as well just at a slightly later date.
I haven’t studied this but are census tracts based solely on property values? Because if there are other elements that enter into the geographic process that will contaminate the data for RP values right there. I may well use Census Tracts as one tool in the tool box but not to be forced into a mold that may look good on paper but really in many cases has little significance to the actual project. If I am forced to use that data at the risk of my license I am forced to lie about reality. If the loan belly’s up three years from now you can bet that the lender will disavow any responsibility they may have created based on their insistence that I had to use their CU or suffer potential blacklisting. Now, dammit, I’m boots on the ground and I know what I’m doing period. They’re not paying enough to have me spend a lot of extra time getting into a protracted discussion of why they think their comp is better than mine on every appraisal.
I love my job but I no longer have the option of firing a client when they become a pain in the ass. Sooo….. a job was what I was looking for when I found this on and there’s an old saying”..if you choose to stay and work in an abusive situation, its not the abuser’s fault”.
Jared Michel did an expert job of describing the coming CU fiasco to appraisers. These are his statements from another blog site:
“There is really nothing wrong what Fannie is attempting. But it needs to be acknowledged that it represents a shift in the nature of appraising for the secondary market. The traditional Fannie appraisal has been the Appraiser’s opinion of the value of the property supported by facts and evidence. The facts and evidence have typically consisted of an analysis of several recent sales of similar types of property. Anecdotes essentially, that support the appraiser’s opinion.
What Fannie is contemplating is asking the Appraiser to complete a more holistic review of the market. Instead of simply presenting data that supports our opinion, we will be asked to also refute other data that might be construed to indicate a different value. Nothing wrong with that, but it should be obvious that this is a much bigger job as there is rarely any logical end point to the task of “proving a negative. Expect prices for appraisals to rise.
If I ask you to describe the taste of a glass of wine you could do so with a couple of simple words…”dry, fruity, hints of oak and chocolate”. If I ask you to explain what that same glass of wine _does not_ taste like its a lot harder job…”It doesn’t taste like beef, or pork, or chicken, or nachos… or cake…, or scotch,… definitely not brussell sprouts or kale…. You get the idea.”
I learned this lesson years ago when our largest client attempted to pull the same stunt on us by sending us 20+ comparable sales. We promptly responded that we were asked for “OUR” opinion of value however we would be happy to analyze other comparable sales and listings as another assignment for $50 per comparable sale. We never heard from this lender again (nor did we wish to hear from them again).
You’re in for a real treat appraisers.
More pertinent observations Dave! Good job.
I would add that it does not appear anywhere in the process that the entire appraisal report as composed by the appraiser, is reviewed either by computer or human.
Now SUMMARIES of our work are being reviewed in a system that was not able to do so accurately in the first place. Many of us expressed concerns when xml was first proposed. THIS is one reason why!
Reports that have had data ‘sanitized’ or excluded are being used to help make collateral related lending decisions. Where is the compliance with SR2?