FHFA Request opinions from Lenders & others on what the new version of an appraisal will be.

[CGinMN]

Face the facts. Appraisers could never keep up with the historic refi demand that was created in part by waiving appraisals.

 

[Mike Kennedy]

NEW Appraisal Form 1 half page plus map and photo pages >>

$$$ ___________ here is the Bull$eye hit it or you're history

 

[Terrel L. Shields]

Appraisers could never keep up with the historic refi demand that was created in part by waiving appraisals.

That's debatable. If banks can keep up with it, why can't the secondary market? After all, they are about equal in size. Why would the evaluation style reports not be acceptable in secondary market but hybrids are? 36% of the loans are held by community banks. So how do small banks have a lower loan-loss rate than the big boys?
From the 2020 FDIC report.

 

[CGinMN]

Face the facts. Appraisers could never keep up with the historic refi demand that was created in part by waiving appraisals.

Emphasis added. Bad sarcasm I guess.

 

[Mejappz]

Appraisers can keep up with the current demand if they are not forced to use the existing forms and reporting requirements, and the scope of work ramrodded down their throats by the client/intended user. Changing the wording and certifications/scope on a 1004 is putting lipstick on a pig. Once the housing market crashes, the GSEs will be fully responsible.

 

[Sadie]

The market has been used to sub-standard appraisals and so they have become the norm: Funky floor plans with only the names of rooms written in, exterior only measurements that always seem to result in the exact same GLA as the tax assessor values, no explanations for anomalies. - Now that users are starting to get something better - and something worse, they will start demanding more of what they want.

This doesn't sound like anything I have experienced. Maybe you just have a low quality sample set?

The story here is simply that appraisal as a profession is changing - just like a lot of other professions.

Less expensive homes that fit the norm may do just fine with AVMs for many loans. Expensive Custom Luxury homes, which tend to meet specific functional requirements will often require more work. Again, if there is a lot of equity in the home, i.e. a low LTV, then the appraisal is simply not that important.

 

[Terrel L. Shields]

photographing the entire house

Which assumes that the house will never change???

Computerized "values" are going to reinforce computerized values. The HO will have hell trying to "correct" the computer, even in the face of substantial changes made. This is a system of fixing values for all properties and only adjusting on the basis of "market conditions."

 

[RCA]

This doesn't sound like anything I have experienced. Maybe you just have a low quality sample set?

How exactly does the typical appraiser get their adjustment for condition, quality, view, functional utility and other intangible features? Just tell me exactly how objectively they go about doing this, and that should answer your question, because as far as I have seen, all across the board from the residential to the MAI who does skyscrapers, it is largely subjective guesswork. When you start adding up these adjustments for the intangibles, then the impact of any errors multiply. The intangible area is the place where appraisers can inject bias, inflate or deflate values based on their holier than thou "experience."

 

[Sadie]

How exactly does the typical appraiser get their adjustment for condition, quality, view, functional utility and other intangible features? Just tell me exactly how objectively they go about doing this, and that should answer your question, because as far as I have seen, all across the board from the residential to the MAI who does skyscrapers, it is largely subjective guesswork. When you start adding up these adjustments for the intangibles, then the impact of any errors multiply. The intangible area is the place where appraisers can inject bias, inflate or deflate values based on their holier than thou "experience."

Silly you. Of course quality and condition are subjective. I think even USPAP addresses that. You cannot make a quality or condition determination without being subjective. However, trained eyes on a property is better than no eyes or untrained eyes. A trained appraiser is going to be better at estimating quality and condition than a algorithm. How does an algorithm account for the subjectives and or the intangibles? Ignore them? Pretend they don't exist? Thats not a good solution. Remember, someone inputs the data and parameters. Who is checking the data input and the veracity? Who is making sure there is no bias in the algorithms? There is room for bias in everything.

 

[RCA]

Silly you. Of course quality and condition are subjective. I think even USPAP addresses that. You cannot make a quality or condition determination without being subjective. However, trained eyes on a property is better than no eyes or untrained eyes. A trained appraiser is going to be better at estimating quality and condition than a algorithm. How does an algorithm account for the subjectives and or the intangibles? Ignore them? Pretend they don't exist? Thats not a good solution. Remember, someone inputs the data and parameters. Who is checking the data input and the veracity? Who is making sure there is no bias in the algorithms? There is room for bias in everything.

Actually you are wrong. They can be made very objective.

In simple terms if you have a good regression model on the tangible features, you get an estimated value for the tangible features. Subtract that from the sale price of the comparable and Voila! - you have the total value of all the other features not tangible, i.e. the intangible features. It is completely objective (on the comparable side).

You might ask, what good is the total value of the intangible features? Well, you can estimate the same total for the subject (which is the only thing that does require some subjective input) and subtract that from the value of the intangibles for the comparable, to get the total intangible adjustment for each of the comparables. That's a big difference, because one side of that equation for 6-12 comps, is completely objective and you only focus on estimating the value of the comparable for the subject --- through a specific procedure which minimizes error on that side [remember the regression model does give an estimate of the tangible features for the subject property as well].

Now, you might say, OK, you have an adjustment through this method for the total of all intangible features for all comps, but you still don't have the adjustments for specific intangible features. Well, it doesn't matter, because you can divide the total anyway you want between Quality, Condition, Functional Utility, ...., and it won't have any impact on the adjusted sales price for the comps. That's simple math. However, you do split the intangible sub-totals up, to explain to the user why the adjustments are why they are. And, you do this by studying of course the features of the sales comparables in relation to the subject. There is this underlying constraint, call it the "containment", all of your adjusted values of the intangible features have to total that sub-total you calculated. And if the the sales price was actually lower or higher than you thought it should have been, sorry about that!! This is a very tough constraint.

If you execute this method (IVCA: Intangible Value Containment Approach) correctly, you are guaranteed to pretty much have all adjusted sales prices within 1% of their average.

What can go wrong? Well, if you don't have a good regression model, you will likely find it difficult to split the sub-totals in a way that makes sense.

Here is an old demo, I actually have newer appraisals that are better, but this should do:

Sample Valuation - Intangible Value Containment Approach

This is a heavily redacted report that was created using the Subjective Value Containment Approach

pvn.pub

Another way of putting this is that with IVCA, the Sale Price of the comparable comes first; while by the traditional approach, the appraiser first cooks up an adjustment for the comparable out of thin air (so to say). The IVCA is truly geared to actual Market Value, the Traditional SCA is geared to the appraiser's subjective imagination and is totally *** backwards.