3.6 The promises, the predictions, the panic and the fight for the dwindling appraiser dollars

[Terrel L. Shields]

Residential fees? can you do 2 per day making $500-800? 2 reports. 2 liabilities to carry until the Statutes of Limitations expires.

Chicken farm? Takes 3 days to do. There are only X many sales in a given area, not 100. X many types (broiler, pullet, grandparent, commercial egg, breeder egg) So you may have only 3 to maybe 6 sales to examine and you get to use them over and over. Current fee? $4,000 absolute minimum up to $6,000 plus.

Vacant commercial warehouse? larger ones - probably a 2 day job, fee $2,000 or so here. vacant farm land is dependent upon data available and $2,000 is not a common fee here but in farm country seems to be typical. RV Parks and Marinas? $3-6k. They are doable in 3 to 5 days depending if you have any prior sales already analyzed.

 

[George Hatch]

So there are appraisers out there that don't go into the MLS, narrow sales down on their own, and then select the best 3-4 sales on their own? Then take one individually, sort through all the discrepancies in the MLS and enter them into the grid? They have a computer pick their comps and import all the "data"?

If I were an appraisalware vendor I would list them all (maybe 40 or 50 of them), highlight and pre-load the ones that look more similar based on certain criteria but leave the final comp selection to the appraiser. They could replace any or all from the larger group or could manually input a transaction if it didn't show up on the list to begin with.

TBH, I'd be a little surprised if at least one of these vendors haven't already built that convenience into their program. Their box chooses 4 or 5 sales but the appraiser is free to disagree and choose to replace them all with his/hers/theirs choice in a 2-minute iteration process. If they haven't done that yet then it's just a matter of time before they do.

What you might be worried about is appraisers who arbitrarily just agree with the initial selection without take the effort to make their agreement or disagreement based on their own discretion. Which some appraisers already do when they run a single MLS query, sort by price and present the 3 highest priced transactions for their SC grid. Same mindlessness in both except one is expedited by use of the appraisal report app.

 

[Alsie35]

What a clownshow. (CRSDATA also pulls in 15 it thinks are most similar for me to look at) but none of these newfangled programs can do what my spreadsheet can do. It allows me to, at single click, order ALL the market area comps, up to 2500, by any criteria, instantly, including by subdivision. It also calculates the 1004MC, and accurately, based on the price range parameters I input for the calculation.

The comps are all spread out gloriously in front of me on my big screen, allowing me to instantly rearrange and compare them by any criteria I want. Then I narrow it down, repeat, and after I've made the final selections, I type them into the grid manually.

Allowing some program to suggest some number of comparables ensures, with 100% certainty, that you're going to miss comps that you might need to bracket something or other.

For example, I can highlight, click copy and paste in one move, all of the MLS numbers from my selected comps, put them into the MLS, and instantly see if any of them share any location factors that I need to bracket on the MLS map.

My spreadsheet was last iterated in 2011. And it is STILL unbeatable. I managed to create a picklist for it on SABOR, an MLS system it was not originally designed to work with, before we moved to TX. The functionality actually improved! It's been the backbone of our entire appraising business for 24 years, including its earlier iterations. It was originally designed by a former retired USArmy 0-5 tech wiz whom I knew at WSB and who was one of my trainers. It was so useful that WF adopted it and iterated it several times. I'll wager that NONE of these 'idiot-we-select-the-comps-and-input-them-for-you' programs can hold a candle to it. The ease of this system is what allowed us to be so efficient.

But that's not the point anymore is it. They now actively want for software to replace you. Your judgment, honed over decades, be damned.

Like I said, what a clown show. Glad I don't need to be a part of that world.

 

[George Hatch]

All that means is that the concept of auto-enabled data wrangling and pre-population has been around for a long time. Including your own app. And that the EASY button apps like Data Master never shut the appraiser out of the comp selection or analysis process. Including overriding any factual errors in the data itself.

Serious question: Regardless of the GSEs long term intentions, do you think the 3.6 apps from ACI or Total are going to limit an appraiser's discretion in comp selection or analysis? If the answer is "no, I don't think that" then all this fretting about appraisers working for the machine instead of the machine working for the appraiser is silly. Downright silly.

 

[J Grant]

All that means is that the concept of auto-enabled data wrangling population has been around for a long time. And that the EASY button apps never shut the appraiser out of the comp selection or analysis process. Including overriding any factual errors in the data itself.

The problem is the churn and burn appraisers will use the EASY button and the other appraisres will not.

 

[George Hatch]

True. And at the same time that is not an argument for limiting the development of this technology. Texas Instruments is not responsible for how I use the HP-12c I bought from them. Microsoft is not responsible for how I use MS Word or Excel. Apex is not responsible for the accuracy of my diagrams.

Taking the (anti-donkey) argument to it's logical conclusion, if the only way to prevent them from abusing the EASY button is to take it away from everyone then that's going to be a tough sell.

 

[Alsie35]

 

[glenn walker]

The churn and burn LPA will always select comparables by highest sale prices to make the target. His bad behavior won't be effected by how he receives his comparables. The machine just increases his speed at hitting his target value.

 

[J Grant]

If I were an appraisalware vendor I would list them all (maybe 40 or 50 of them), highlight and pre-load the ones that look more similar based on certain criteria but leave the final comp selection to the appraiser. They could replace any or all from the larger group or could manually input a transaction if it didn't show up on the list to begin with.

TBH, I'd be a little surprised if at least one of these vendors haven't already built that convenience into their program. Their box chooses 4 or 5 sales but the appraiser is free to disagree and choose to replace them all with his/hers/theirs choice in a 2-minute iteration process. If they haven't done that yet then it's just a matter of time before they do.

What you might be worried about is appraisers who arbitrarily just agree with the initial selection without take the effort to make their agreement or disagreement based on their own discretion. Which some appraisers already do when they run a single MLS query, sort by price and present the 3 highest priced transactions for their SC grid. Same mindlessness in both except one is expedited by use of the appraisal report app.

Fannie Mae is on a journey of continuous improvement to make the home valuation process more efficient and accurate. We provide a spectrum of options to establish a property’s market value, with the option matching the risk of the collateral and the loan transaction. The spectrum balances traditional appraisals with appraisal alternatives.

Value acceptance is Fannie Mae’s offer to accept the lender-submitted property value in Desktop Underwriter® (DU®) with no appraisal requirement for eligible transactions. The value submitted must be based on the contract price for a purchase transaction or the lender’s or borrower’s estimate of property value for a refinance transaction.
My comment - the lender or borrower or contract price did not even need a robo program pick of 3 comps
Having a lender pick the value - wasn't that what the HVCC and Dodd Frank seek to eliminate via stopping mortgage lender pressure to hit values in an appraisal! The GSEs had the idea to cut out the appraisal and thus the regulations around appraisal predetermining value, to allow the lender to estimate the value to hit X number. The HVCC and third-party AMC buffer against value hitting in appraisals - the GSE solution was not to use an appraisal, and regulators went along with it.

When this is what appraisals compete against on the res lending side, no wonder some shortcut appraisers use a software program to pick comps. I won't do it and others won't do it, but others will, and who is going to stop them? The same folks who decided dropping appraisals was fine in order to allow a lender to estimate the value?

​

Fannie Mae
https://selling-guide.fanniemae.com › sel › representatio...


Fannie Mae does not warrant that the estimated value provided by the lender is the actual value of the subject property. The lender may not make any statements ...Read more

 

[J Grant]

True. And at the same time that is not an argument for limiting the development of this technology. Texas Instruments is not responsible for how I use the HP-12c I bought from them. Microsoft is not responsible for how I use MS Word or Excel. Apex is not responsible for the accuracy of my diagrams.

Taking the (anti-donkey) argument to it's logical conclusion, if the only way to prevent them from abusing the EASY button is to take it away from everyone then that's going to be a tough sell.

If the stakeholders were serious about appraiser standards, then they would take the EASY button away from everyone (by not allowing its use on their orders )