Fannie and Freddie - UMDP / UAD Forms Redesign Initiative - They ARE interested in appraiser input. Surprise!

[djd09]

Does Affirmative Action in Mortgage Lending Really Help Black Americans?​

There’s no clear reason to believe that race-conscious lending benefits those whom it’s meant to.​

The potential for taxpayer exposure to loans made for less-than-financially sound reasons was made clear by the $187 billion federal bailout of Fannie Mae and Freddie Mac in the wake of the 2008 financial crisis,

We’ve seen this movie before. The 2008 financial crisis and its widespread mortgage foreclosures disproportionately affected black Americans. Some 240,000 black homeowners lost their houses in the crisis, and home equity among their neighbors plummeted. In predominantly black and affluent Prince George’s County, Md., the median home price dropped from $343,000 to just $245,000 between 2008 and 2009.

the corrupt leading the way, it will work

Welcome to banking’s ‘reaction season’​

Over the span of a few hours, three big-bank CEOs offered counters to running narratives on mortgage lending, shareholder proposals and a “hurricane” prediction.

Wells Fargo CEO Charlie Scharf offered a striking example Wednesday, addressing a mortgage-lending morass that has enveloped the bank for more than two months. A Bloomberg analysis in March found Wells Fargo approved just 47% of mortgage refinancing applications filed by Black homeowners in 2020 — a percentage that far lags its competitors.


Scharf on Wednesday indicated that, to some degree, Wells’ hands are tied.

“We basically process the applications according to guidelines that [government-sponsored entities like Fannie Mae and Freddie Mac] tell us we should,” he said, according to Business Insider. “When those produce results, the people like them, we get the kudos for it. If they don’t like them, we get the blame for it even though we’re just following other people’s underwriting guidelines.”

Welcome to banking's 'reaction season'

Over the span of a few hours, three big-bank CEOs offered counters to running narratives on mortgage lending, shareholder proposals and a "hurricane" prediction.

www.bankingdive.com

you know what i can relate to that statement. the user abusers want this and that, don't like the results, and then proceed to whine and cry

 

[Terrel L. Shields]

the user abusers want this and that, don't like the results, and the proceed to whine and cry

Exactly. Poor people get turned down more often than rich. Blacks and Hispanics tend to be poorer than whites. Sheer logic dictates that a higher rejection rate will apply. If the government don't like it, then they should make those loans themselves and let the taxpayers absorb the losses...oh, they already do that.

 

[RCA]

for some of you kids on the block. soon after the 1004 was in constant use, they added the 2055 to make it faster. nobody really used it then. but they then came out with a 1 page 'quantitative' appraisal form where all you had to do was put a '+' or '-' minus sign, then give a value. that went no where, but the circle of appraisal life is very strange. we are now trending to the minutia data appraisal, or an observation home inspection value report. i agree with data mining for fannie mae's computer model. someone there thinks that this 'more' detailed form will help their fannie avm to give a 'more accurate' value. the government doesn't care about people. it just happens that sometimes you think they like you, god forbid they don't. my opinion has been that certification created too many appraisers. i can survive any upcoming apocalypse, have lived thru enough of them, although some barely.

Most people who buy a house are buying it for a local, relatively small geographic area. Thus all pricing is relative to local geographical areas. As a sign of how utterly stupid FNMA and Freddie Mac are - they still haven't figured this out several decades into the game. They are still stuck with the absolute standards C1-C6, Q1-Q6 for the entire United States, as if people up and said to themselves: "Where can I find the best deal on a house in the United States?".

The solution of course is simply to go with percentage better/worse than the sales comparables that have sold in the past X months for your geographical area ---- whatever you choose, I use - 0.00 to 10.00 for 0.0- 100.0%. Why such accuracy? Well, because price curves are very steep near the low and high end. I can rate my comps with residuals and maybe I see two houses A and B. A's score is 0.5 (it is better than 5% of homes sold) and B's score is 0.6 (B is better than 6% of homes sold) my comp is right between the two. So, I give it a score of 0.55. This will give me an estimated residual, which added onto the regression model estimate based on measured features - gives me the value conclusion.

So easy.

Well FNMA and Freddie Mac I guess cannot be called idiots - because as far as I know I am the only appraiser in the world so far smart enough to have figured this out.

But considering their size and the money at their disposal, I will go ahead and call them for what they are: Worse than idiots.

 

[George Hatch]

One of their considerations is that whatever rating system they come up will be used by appraisers *and users* of varying competency and effort. The quality and condition levels are benchmarked and observable, whereas leaving those ratings to the discretion of individual appraisers based on other properties in the area is moving away from that. "Inferior, similar, superior" does work insofar as the one dataset being used in the one appraisal at hand, but it's not going to work as well when a user is engaged in analyses of larger datasets.

MAYBE we should consider the optimum solution to include "both" as opposed to assuming it has to be an either/or choice.

We write appraisal reports for our users to use, not for ourselves or to assist us in developing our opinions. The forms (and the format) don't drive the analysis, they communicate them. Ideally in a manner the users can understand.

Appeal to reason (what makes sense to the reader) > appeal to authority (trust me because I'm the appraiser). IMO

 

[George Hatch]

I wonder if a Q/C checklist could be developed which would be used to get to more consistent overall ratings. Maybe use it to rate several individual components for their respective Q and C ratings. Maybe even assign a weight to each component so the score could reflect the weighed average of the whole.

Three columns; component, quality, condition.

 

[Carnivore]

GH, I agree with you above. We already Condition rate Kitchen and Baths. and indicated when they were last updated. The GSE's may have now included a Q Rating for those same areas.

Case in point. My last assignment (conventional) The owners are elderly, they renovated the Kitchen baths a long time ago. Owner(s) couldn't remember when they did it very well. What. I observed is the elderly couple are very easy on the interior. Frankly if she had told me about 2-3 years ago I would have believed her. Quality was easy , no doubt Q4. Formica Tops, Fiberglass tub enclosures. Standard Kitchen cabinets etc Roof shingles were standard 25 year and they looked to be about 10 years The SH Vinyl DP Sash Q4. They looked new but I knew better but they were not sure . The owners are clean freaks.

 

[Terrel L. Shields]

Maybe use it to rate several individual components for their respective Q and C ratings. Maybe even assign a weight to each component so the score could reflect the weighed average of the whole.

I actually do use a comparative table of comps and the subject on some property, particular unique or very high end stuff.

 

[Non Sequitur]

We can't account for the fall of every sparrow in GSE reports. Describe in plain english the subject's salient features and explain how the comparables compare in your analysis and reconciliation. The idea that checkboxes and pull-down menus are enough is leading us down the path to the death of our profession.

 

[Mejappz]

Rating different rooms, kitchens, etc., will do nothing to improve valuations. Bert is right; the C & Q rating are a joke. I suggested something similar to what Bert suggested and got laughed at by the appraiser Illuminati. We will see who gets the last laugh.

 

[George Hatch]

For the last 10 years I've always run a grid showing ceiling heights and floor coverings, and another grid showing ratings for major factors, albeit not for specifics like floor coverings or paint or whatnot. It takes almost no time to collect that info and only a couple minutes more to enter the info into my report. Doing so actually *reduces* the amount of original writing I do to summarize the property attributes. Although I do that, too, for both the composition and the condition.