“(Bob) Murphy (Fannie Mae) does acknowledge that Fannie Mae is able to track photos in each appraisal, a practice many appraisers have long suspected, which means that Fannie Mae is able to detect when appraisers reuse comparable photos in different appraisals and flag appraisals which contain outdated photos as deficient.”
Another “good one” by Dave Towne, Washington appraiser and commentator
Something’s been gnawing at my craw ever since January when FNMA’s wonderful CU was unleased to the world. And before that, which still continues, is the AQM process they still use to judge the work of appraisers.
No one else has written about this, or even mentioned it, so I will: It has to do with the word “Comp” which is used liberally by FNMA.
What exactly is a “Comp?”
In FNMA’s world, it’s any property that they obtain, either by their vast AVM process which examines millions of property transactions, or properties that have been extracted from appraisal reports submitted by appraisers……..yes, your work. In their fuzzy logic, it’s a “Comp” considered for your report if they say it is. It is not!
A true “Comp” is a property viewed and/or analyzed by a real living, breathing, mirror fogging appraiser who compares that sold property against the subject property in terms of multiple features, characteristics and amenities. It is not determined by an AVM or algorithm within the vast bowels of FNMA. Until the property has such analysis done by an appraiser, it is merely a SALE……it is not a “Comp.”
This FNMA lie really became evident to me on 4/20/15 when FNMA released a news release about how CU has been integrated into their on-line Desktop Underwriter software mortgage lenders use, which you can read here: http://www.fanniemae.com/portal/about-us/media/corporate-news/2015/6239.html?p=Media&s=News+Releases&from=RSS
Within that news release is this quotation from a VP at a mortgage lender: “The collateral information that CU provides is invaluable and simply staggering,” said Breck Tyler, Executive Vice President, Trustmark Mortgage Services. “CU has aided in providing important comparable data that was previously unavailable or very difficult to get. CU messages in DU will help streamline appraisal review and make the underwriting of an appraisal a much more informed process.”
Then, FNMA released info directed to Correspondent Lenders who intend to use the CU process in UCDP, but don’t intend to sell the loan to FNMA: https://www.fanniemae.com/content/fact_sheet/collateral-underwriter-non-seller-implementation-guide.pdf
That has this statement: “Fannie Mae does not instruct or suggest to lenders that they ask appraisers to address all or any of the up to 20 comparables that are provided by CU for most appraisals.”
I want to repeat what I said above…in case you missed the point: A PROPERTY IS NOT A “COMP” UNLESS YOU DETERMINE IT IS AND INCLUDE IT IN AN APPRAISAL REPORT. Otherwise it’s just a ‘sale.’
If you’re an appraiser who liberally uses the word “Comp” in place of a ‘property sale’ I would ask that you be more careful. If you receive info from a lender, AMC or anyone else who asks you to look at the “Comp” they have provided, correct them and use the words “sale property” until you have determined that it truly is a “Comp.”
I’m also asking members of appraisal organizations and associations to communicate your concern about this lie perpetrated by FNMA directly with them, and ask FNMA to change the word “Comp” used in their CU Reports, news releases, instructional materials, etc. to ‘Property Sales’ so that there is no misunderstanding about the significance of this issue.
If organizations and associations won’t do that on behalf of appraisers, then we might as well kiss the profession of residential real property appraising goodbye. Because if a list of ‘sales’ are considered “Comps” then an actual human appraiser won’t be needed to provide supportable property analysis and market value reports.
My comment: I have been aware of the difference between a sale and a “comp” for a very long time. I try not to mix them up. It is very important when communicating with lenders and real estate agents, who should already know the difference. I am glad that Dave Towne points out this very big difference.
I have not found it to be an issue with non-lender clients, where I use “comparable sales” which is a much clearer term to use, since few are familiar with the term “comp”.
A disgruntled man who lost his home to foreclosure did it!! The CU system is completely shut down. No details were available on when, or if, it will be fixed. The un-named man was not available for comment as he is reportedly hiding out in a remote cabin in the woods in a very rural upper Michigan or Canadian location.
He, his wife, and 4 children are now living in a Tiny House on his brother in law’s rural property in Michigan. They lost their suburban 4 bedroom home they had owned for over 20 years to foreclosure. The family has an outhouse, a hand pumped well and wood heat. A small solar power array is used for a computer and Internet connection with a few hours of electricity daily.
Who is in an uproar about this?
– Not appraisers who don’t like a system that evaluates their appraisals but they don’t know the criteria.
– Not underwriters who are required to learn a new system and do manual reviews of appraisals before sending warnings to appraisers.
– Not loan officers whose deals are taking longer to close. Not borrowers who have to wait longer to get their loans approved.
– Not law enforcement, who don’t think he is a terrorist or a nut holed up with a lot of guns.
FYI, it is April 1 today. You know what that means ;>
The key to any statistical analysis is DATA, DATA, DATA!!
Single family real estate data is not very reliable or consistent, and not enough is available in many areas, as we all know.
CU is the most significant attempt to get more useful data by requiring appraisers do use specific coding and criteria. However, real estate is local, local, local. Even the number of bedrooms varies a lot as there are different criteria for determining what is a bedroom, even in the same city. Three appraisers measuring the same house will probably not have the same square footage, as I learned doing relocation appraisals.
With CU, this is becoming more obvious as there are sometimes wide variations in how appraisers code factors. For example, why do condition ratings vary? How accurate is MLS? Is public records accurate? What is the best source?
Now that regression software is popular with appraisers for getting adjustments, I have been thinking about why it is often not very reliable. To understand even simple regression requires knowledge.
My first statistics class was in 1963. The first time I used multiple regression was in graduate business school in 1979, when I did a mini-thesis on factors in REIT stock volatility using SPSS.I used a remote university mainframe that kept blowing up and erasing my data. There were no data issues. Doing multiple regression analysis on real estate housing data was not possible. Way too much lack of usable data.
Since I started my Appraisal Today newsletter in 1992, I have been writing about AVMs. The less data that is available, the less reliable the value.
As we all know, AVMs work well in a conforming home in a large tract of similar homes, built in the past 10 years. After that, the accuracy and reliability goes down fast. Just check what Zillow’s Zestimate against your appraised value.
Fannie is using this to show that appraisers have been using adjustments that are too low, resulting in less reliable values. They are often low “legacy” adjustments. Also, GLA adjustment is one of the few factors that work well in regression.
I suggest using replacement cost new less depreciation. For replacement cost you can use local builders or cost service such as Marshall & Swift, whichever is more accurate in you area. Then take off depreciation. The result is depreciated cost. Divide by GLA. The result is depreciated cost per sq.ft.
Fannie uses price divided by sq.ft. which does not consider land value or depreciation, information which Fannie does not have available.
For example, builders cost on a property is $100 per sq.ft. Your estimated physical depreciation is 30%. Obviously, $25 per sq.ft. adjustment is not correct. There may be functional or external depreciation, which you can include. Be sure to include how you determined your GLA adjustment in your appraisal.
Market based GLA adjustments are better, such as matched paired sales but the method above will work as a guideline.
Why are adjustments low? To comply with the 15/25% adjustment guideline, which Fannie has removed. It was never a requirement. Fannie has never had a 10% per line adjustment guideline. Of course lenders and AMCs can still require the use of the 15/25% adjustment which could be a big problem for appraisers which can result in less reliable values. I never considered the 15/25 guideline in any of my appraisals, but I never worked for lenders or AMCs who required that appraisals conform to it.
Check out the graphs on GLA and 15%/25% adjustments in the FAQ document below. I included 4 of them in this month’s paid Appraisal Today newsletter.
Get the facts about what Fannie is saying, not just rumor and speculation. Subscribe to the paid Appraisal Today!!
My latest opinions and observations, as of today
Fannie does not want appraisers to receive warning messages unless a “human” has reviewed the appraisal report. They want to reassure real estate agents mostly that appraisals will not be delayed. Of course, I have no idea how many underwriters have the time to read the 30+ page report. Maybe they can search the report for what they are looking. I am sure this is/will be slowing down loans.
But, I keep thinking that even if appraisers received a few CU warning messages, it is a small, small percent of all the stips from all the review software that AMCs use. No one seems to notice that appraisals take longer the more stips that appraisers receive. Particularly, when all the stips are not sent at the same time. No one seems to notice this, or care about it, except appraisers!!
These non-CU stips are mostly from arbitrary “rules” which CU does not use. Such as: picky UAD stips, “add 2 more comps”, or please review the list of “comps” from the real estate agent or borrower. Some are still using the 15%/25% adjustment rule.
Fannie says that typically there are around 7 UAD records per property. However, most of them must be from appraisers who used it as a comp. Why should we be compared with appraisers who used the property as a comp?
I want access to CU property data from appraisers who did an appraisal on the subject property, not from appraisers who used it as a comp. I want CU to use this data to compare my appraisal data with “peers”. Comp data is not very reliable as it usually comes from MLS and public records. Fannie says that MLS and public records are not as reliable as data from the appraiser who appraised the property for the sale.
Maybe the appraiser had seen the interior of the comp recently, but this is very unlikely. Also, I go on the MLS tour/caravan almost ever week, but I don’t spend a lot of time at each open house. Well..I do spend more time if there is good food ;> MLS photos are subject to interpretation as they are done for sales purposes. I make brief notes on the flyers and file them in binders, going back to 1990.