Collateral Underwriter "suggested Comparables"

[timd354]

Doppel-Post

 

[Meandering]

Your conclusion is not neccesarily supported by the data. The data could also lead to the conclusion that appraisers artificially manipulated adjustments to fit within the guidelines. Unless one did an in-depth analysis, which I am sure that Fannie did, they could not determine the correct conclusion. Obviously, Fannie has concluded that the data show that adjustments were manipulated and changed their guidelines as a result. Whatever you think of Fannie, I can tell you that the people who analyze appraisal data are very intelligent and very talented and I really doubt that their conclusion is incorrect.

I do not doubt their intelligence.

I doubt their statements as presented with the support of their chart. The chart does not match what they are saying.

If adjustments are "artificially low", and I don't doubt there may be some, then the adjustment, should be reflected in a dollar amount. Like the $600,000 house that had a $5,000 adjustment (<1%), instead of a $50,000 adjustment (8.3%), and Fannie is instead expecting a > $150,000 (25%) adjustment simply because the price of the house is "higher" than the state average.

And that the $60k house with the $5,000 adjustment (8.3%), should have been closer to $500 (<1%), because it is a lower priced home than the $600k home, and lower priced than the state average.

But when you take the median, which is used instead of the average particularly because it does not give weight to the outliers, then, from what they are saying, they expected the median percentages to be higher on higher priced homes, suggesting that the more expensive the homes are,(the MAJORITY, because we are talking median)s, the less comparable the comps should be, causing the percentages to be higher.

 

[Eli]

This has nothing to do with differences in sale prices, because the gross and net are a percentage of the sale price, so the dollar amounts of the gross and net would change with the sale price changes, not the percentages.

.

I do not doubt their intelligence.

I doubt their statements as presented with the support of their chart. The chart does not match what they are saying.

If adjustments are "artificially low", and I don't doubt there may be some, then the adjustment, should be reflected in a dollar amount. Like the $600,000 house that had a $5,000 adjustment (<1%), instead of a $50,000 adjustment (8.3%), and Fannie is instead expecting a > $150,000 (25%) adjustment simply because the price of the house is "higher" than the state average.

And that the $60k house with the $5,000 adjustment (8.3%), should have been closer to $500 (<1%), because it is a lower priced home than the $600k home, and lower priced than the state average.

But when you take the median, which is used instead of the average particularly because it does not give weight to the outliers, then, from what they are saying, they expected the median percentages to be higher on higher priced homes, suggesting that the more expensive the homes are,(the MAJORITY, because we are talking median)s, the less comparable the comps should be, causing the percentages to be higher.

I think they are saying they want adjustments and analysis supported by recognized technigues and methods period. They are sayin forget the percentages or dollar amounts as long as the adjustments and analysis are supported by recognized appraisal techniques and methods.. That's my take. Of course I am not trying to be arrogant because their take is based on indications of unsupported adjustments from their analysis and I am just reading it. So don't be mean to me, cause......I'm just reporting ....LOL

 

[timd354]

I do not doubt their intelligence.

I doubt their statements as presented with the support of their chart. The chart does not match what they are saying.

If adjustments are "artificially low", and I don't doubt there may be some, then the adjustment, should be reflected in a dollar amount. Like the $600,000 house that had a $5,000 adjustment (<1%), instead of a $50,000 adjustment (8.3%), and Fannie is instead expecting a > $150,000 (25%) adjustment simply because the price of the house is "higher" than the state average.

And that the $60k house with the $5,000 adjustment (8.3%), should have been closer to $500 (<1%), because it is a lower priced home than the $600k home, and lower priced than the state average.

But when you take the median, which is used instead of the average particularly because it does not give weight to the outliers, then, from what they are saying, they expected the median percentages to be higher on higher priced homes, suggesting that the more expensive the homes are,(the MAJORITY, because we are talking median)s, the less comparable the comps should be, causing the percentages to be higher.

I have absolutely no idea what you are trying to say in this post Marian as the Net and Gross adjustment charts and discussion included in the Lender Letter have make reference whatsover to loan amount (or include any data re: loan amount) but just show the distribution of gross and net adjustmens on a % basis....where to get from those 2 charts that Fannie is expecting a larger % (or dollar) adjustment based on the the loan amount or state average.

I believe that you may be mixing data from the 2 charts that address the distibution of net and gross adjustments with the data and discussion from the GLA adjustment discussion and charts which follow the discussion of the distribution of net and gross adjustments

The bottom line is that their data show that 94% of comps in all appraisals submitted to the UCDP have had net and gross adjustments that fell within the 15% net/25% gross former guidelines. Apparently, from there internal studies, they believe that number is artificially high due to manipulation of adjustments by appraisers. Without having access to their data, I have no reason not to believe their conclusion, especially since it intuitively makes sense to me that 94% of all comps would not have adjustments that fell within the 15%net/25%gross adjustment parameters if appraisers were not fudgng the adjustments to fit within the box

 

[Meandering]

I think they are saying they want adjustments supported by recognized technigues and methods period. They are sayin forget the percentages or dollar amounts as long as the adjustments are supported by recognized appraisal techniques and methods.. That's my take.

What they said was:
Nearly 95% of comps analyzed had net adjustments less than 15%, suggesting that appraisers strictly adhered to the net adjustments guideline. Fannie Mae’s concern is that the adjustments might be artificially low.

Appraisers also appear to adhere to the 25% gross adjustment guideline, with nearly 94% of comps having gross adjustments less than 25%. Again, this raised concerns with the accuracy of these adjustments

So if net is less than 15% it is pretty rare that gross would be more than 25%. But here it is, the sentence above and below their 1st chart.

http://www.scribd.com/doc/254761393/FNMA-Lender-Letter-LL-2015-02-About-CU#scribd

and,

The next chart shows the median amount of an adjustment for GLA by state. Analysis shows little variation in the median GLA adjustment despite significant variation in price. Only in Hawaii does the median GLA adjustment break $50 per square foot, even though median GLA price exceeds $50 per square foot in all 50 states and is significantly higher in certain market areas.

Again, this is not portrayed by their charts, which use medians, instead of averages, therefore the outliers are not counted, but the obvious trend line of the GLA adjustment in their state price chart is a definite upward trend from the low <$100k to the lower than Hawaii, >$900 homes, from $15 a square to $ 56 a square with the medians flattening between $303k and $320k, which, given the prevalence of new tract homes during the boom years, might just be accurate with the price differences reflected in 2 versus 3 car garages with little differences in GLA. Because the data is not presented by different sized homes, but rather is presented as price. So $303k and $320k might just be the same sized homes, but with some with views, or extra garages or built in pools, therefore the GLA adjustment does not account for any size difference.

They infer that because the majority of comps are below the net adjustment guidelines they created, that appraiser's are wrong and are making low adjustments, even though, the adjustments are within the Guidelines they wanted.

.
.

 

[gregb]

Whatever you think of Fannie, I can tell you that the people who analyze appraisal data are very intelligent and very talented and I really doubt that their conclusion is incorrect.

How much tax payer $ went into bailing out these "very intelligent and very talented" folks?

 

[djd09]

if the 94%-95% were over the 25% guideline they would say that appraisers are not picking the right comparables. the good ole double edge sword. what would be the correct chart per fannie?

 

[djd09]

and 9.5 out of every 10 appraisers should take offence that fannie in a round about way accused appraisers of violating USPAP and not making the correct market reaction adjustments.

 

[Amy Perkins]

Maybe the CU should write the appraisal and I can approve it instead!!

 

[Amy Perkins]

I agree that most appraiser adjust too low because they want to conform to guidelines. However, this is a mistake and can lead to gross under or over valuation in complex markets.