CU warning messages – grrrr

A few appraisers are reporting getting CU appraisal warning messages from AMCs. Some AMCs get the messages and and some don’t, depending on the agreement with their lender client.

I sorta believed all the “experts” who said CU would not affect appraisers much, except the many us who do not have market based adjustment support in our work files (which we should have always had). “They” said appraisers’ time for responding to AMC questions will not change. Fannie’s reviewers have been using CU for about two years. Some lenders beta tested it. They all liked it. But, I wonder if it was tested with “boots on the ground” appraisers who actually had to respond to the warnings??

In January I wrote up a long CU article for my paid Appraisal Today newsletter. In the February issue I will have another long article, focusing on the differences between the old and new CU warning messages. They are very different. AMCs with access to lender’s warning messages are sending them to appraisers, such as:

Old message (pre-CU): Condition adjustment for comparable property #<comparable number> appears excessive.
New message(CU): The condition adjustment [for comp #X] is smaller than peer and model adjustments
New (CU): The condition adjustment [for comp #X] is larger than peer and model adjustments.

There are other messages about condition ratings different that peers and model.

I don’t know how our “peers” and The Model made their adjustments or ratings and what they are. I don’t know how to respond as to why mine differ.

Now that appraisers are getting the warnings, they are asking how to respond to them. Who are these peers? What is the model? I have no idea how to respond, except to say “I don’t know who the peers are and how they determined condition or what method they used for their adjustment. I am unable to respond.” How do you know what the condition is really like for comps? There are lots of ways to estimate an adjustment for condition. You can explain what you did. But, who is right? You, peers, or model?

MLS is soo reliable (Not) for estimating comp condition. I don’t think they will like “matched paired sales” on all of your responses for the method you used for adjustments.

Looks like maybe there will have to be some webinars for appraisers, not just underwriters, explaining how to respond.

Appraisal Today newsletter

Appraisers not getting access to their own data on CU

There is a petition and a letter being circulated about appraisers getting access to CU, particularly the Web interface which lists comps. This is unlikely for many reasons, which I write about in my paid newsletter.

More important (and more likely to occur) is: Why don’t appraisers get access to subject and comp physical characteristics from the CU database, which was provided by appraisers using UAD?

For example, which appraisers are able to measure their comp GLAs? Not many. This data would really help appraisers do better appraisals. We can always look at MLS interior photos and interview agents, buyers, and sellers for other information we need, such as condition. When the MLS listing says “contractor special” or “fixer” that is a good indicator of condition.

The only reason I have heard is that appraisers vary widely and there are too many differences. GLA is a good example. This has has always varied among appraisers. When I used the old CMDC appraiser database in the late 1980s, sometimes there were more than one source of GLA on a property. I have done relocation appraisals since 1986. It was very seldom that the 2 or 3 appraisers have the same GLA. The “rule of thumb” was up to a 5% difference in GLA was ok.

How many appraisers are “fudging” their dimensions to make their GLA match public records and avoid “stips”? Hopefully, CU will change this. Maybe CU will notice how many appraisers just use public records and how many use their own measurements.

I am really hoping that Fannie allows appraisers to get property characteristic information. It will help all of us – Fannie, lenders, AMCs, appraisers, reviewers, etc.

Appraisal Today newsletter

CU – census tracts, adjustments, "bad apples", etc.

There is a lot of misinformation about CU. No one knows what will happen when CU is fully implemented. I speculate myself. I am an appraiser. I have opinions ;>

UAD is mechanical. CU is asking appraisers to think about their appraisals, not how to classify a characteristic.

For the appraisal profession, I think CU will make us better appraisers by making us take a critical look at adjustments. It will also help get rid of the “bad apples”, including appraisers that “push” values, throw anything into the form to get it out the door, need lots more training and education, etc.

I think Fannie’s main purpose of CU may be to stop appraisers from having low (or high) adjustments, inappropriate comps, using Q/C ratings, etc. to make values higher. That is what they worry about.

Only using comps from within the subject’s census tract is ridiculous and I’m sure CU will not be doing this. It is a good idea to see which census tracts match the neighborhood boundaries that you use. Or, part of Census Tracts. Then you can put the census tracts you use in your appraisal. In some areas census tracts are way out of date due to new construction, plus other problems.

To find census tracts near any property, go to http://www.huduser.org/qct/qctmap.html and type in an address.

I started my business in 1986 and had to put census tract numbers in my appraisals for the first time. I had previously worked for an assessor’s office and had never done a lender appraisal. I used Thomas Brothers Census Tract books to find them. To me, they often represented a reasonable way to delineate all, or part of, a neighborhood. Looking at the current census map for Alameda, CA, my city (population 75,000), it definitely did a good job of defining neighborhoods. However, I usually have to include more than one census tract as there is not enough data to do an appraisal otherwise. It did miss one very important neighborhood where most of Alameda’s large historic homes are located. There is a significant premium for being in this neighborhood. I very, very seldom go out of this neighborhood for comps. I suspect there are issues like this in other geographic areas. I have no idea what area Fannie would use, so I would put an explanation in my appraisal.

The problem is the forms, which were developed for use on tract homes. If you are not appraising a conforming tract home, it is like trying to put square boxes into round holes.

Every appraisal will have a risk score. A high risk score (1.0 to 5.0, where 5.0 is high risk) does not mean an appraisal is “bad”. It may be in an area of declining values or have a negative location problem. Or, not enough comps to provide a reliable value.

Remember that only certain UAD items will be considered by CU for now. If it is not UAD formatted, it will not be looked at. I don’t think Fannie’ use of census tracts will be the issue.

The Big Issue is support for adjustments. I have no idea how to support all the adjustments I make in my appraisals. I know what buyers will pay more, or less, for. But, I don’t know the exact dollar amount.

Regression is just one way to support adjustments, but it will not work for many adjustments, particularly if there are very few sales. Regression is not the only answer. There are many other methods. I will be writing about them in my paid email newsletters.

Regression works very well for time adjustments. Be sure yours are market based, not just from an MC form.

I am seriously considering not making any dollar adjustments when I use form reports for non-lending work, except time adjustments. I never make dollar adjustments on narratives and apartment form reports. My state regulator wants to see support in my files for adjustments.

Just because there is a box does not mean it has to be filled in. Qualitative adjustments are fine. There was a Fannie form developed and used for awhile in the 80s or early 90s that did not use dollar adjustments, only plus or minus signs. I worry about that a lot. The old Fannie 2-4 unit form did not have any adjustment boxes. I really hated when they changed that form to include adjustment boxes and de-emphasize the Income Approach.

No one knows how CU will work out. Will everyone turn down appraisals except for conforming tract homes? Will there be no one to do the tough appraisals and work in rural areas. When appraisers are compared, does the majority opinion win?

Will the days of 24 hour turn times and $200 fees be gone? Will AMCs stop broadcasting all appraisal orders to everyone on their fee panels? Will all appraisers be seen as the same and interchangeable? Or, will appraisers be rated on skills, education and experience? Will fees go up? Will fees be based on difficulty of the appraisal? Will lots of appraisers abandon the lender appraisal ship of fools?

Read the webinar pdfs and look at the maps from the two Fannie Webinars to see what they actually are doing. I spent lots of hours doing this, plus speaking with others about what they thought. Of course, it was for a 12-page article in my paid newsletter. Plus 18 pages of excerpts from Fannie documents and webinars. I probably would not have done it otherwise ;>

Go to www.fanniemae.com/singlefamily/collateral-underwriter and listen to Fannie’s two webinars for underwriters – very good with excellent illustrations and explanations. Plus, read the FAQs. You need to register, but it is very easy and you go directly to the webinar and can return at any time. There are lots of links on the web page for more information.

Last month’s January 2015 issue of the paid Appraisal Today newsletter had a 12-page article on CU plus 18 pages of addenda material. The February and subsequent issues will address problems such as how to make adjustments. Click the ad below for more information.

Appraisal Today newsletter