Blind Squirrel and Acorns

[J Grant]

BPO's...why are they called a Broker price opinion, when an agent does them? And since most pay about $25 now, the successful, experienced agents are not going to do them. Runners take the photos and the desperate or new agents, or front desk person does them while answering phones....( unless the agent lets their HS kid do them for spending money, )....nobody knows who actually DOES them, an agent puts their name on the BPO but no signature is made.

 

[NLCApprMgr]

Agree and Disagree

...there is a general perception that there is an overall upward bias in appraisals (even after the bubble), and believe that most adjustments are not based upon sound statistical analysis of market reaction, but are either just a wild guess or, at best, are usually based on inadequate analysis.

...using a paired sales analysis to derive adjustments is dubious fat best from a statistical perspective, yet this is how most residential appraisers claim to derive most adjustments when what is really needed to derive a an accurate estimate of market reaction to a particular feature would to be conduct a regression analysis using data from many sales.

Residential Appraisers need to up their game or they do face the possibility of losing a business in the future as AVM models improve. From an investor's perspective, there would be absolutely no reason to utilize appraisers if they concluded that collateral risk can be better managed using AVM's. That is not the case today, but it could be the case at some point in the future.

To your first point, I can tell you my experience is that most bankers for non-sale loans generally believe appraisers are misrepresenting values with a lower bias to reduce liability. Personally, I do not agree with either your logic or theirs; the appraiser expresses an opinion based on their view of the market outside the boundaries of a preconceived notion of value.

To your second point regarding the development of adjustments, I agree it would be great if there were a way to efficiently develop a regression analysis on each particular criterion of adjustment; however, the ability to obtain the necessary data so that such a regression were reliable is a fantasy. To wit, how does a regression analysis account for differences in quality or condition? Functional utility? Does a quick deed sale automatically call for a sale condition adjustment because it's not a warranty deed?

To your last point, and based on what I have seen so far, I agree that on a general basis, appraisers need to make sure that they are providing a clear picture of the environment in which the subject property lies, and need to provide more detailed language on how adjustments are developed. There are a number of reports I have seen where the vagueness of language makes me cringe, and the adjustments used by the appraiser might have well come from a Zoltan the Magnificent arcade machine.

 

[ucbruin]

To your first point, I can tell you my experience is that most bankers for non-sale loans generally believe appraisers are misrepresenting values with a lower bias to reduce liability. Personally, I do not agree with either your logic or theirs; the appraiser expresses an opinion based on their view of the market outside the boundaries of a preconceived notion of value.

To your second point regarding the development of adjustments, I agree it would be great if there were a way to efficiently develop a regression analysis on each particular criterion of adjustment; however, the ability to obtain the necessary data so that such a regression were reliable is a fantasy. To wit, how does a regression analysis account for differences in quality or condition? Functional utility? Does a quick deed sale automatically call for a sale condition adjustment because it's not a warranty deed?

To your last point, and based on what I have seen so far, I agree that on a general basis, appraisers need to make sure that they are providing a clear picture of the environment in which the subject property lies, and need to provide more detailed language on how adjustments are developed. There are a number of reports I have seen where the vagueness of language makes me cringe, and the adjustments used by the appraiser might have well come from a Zoltan the Magnificent arcade machine.

NLC, because of your comments, I assume (sorry if I'm wrong) that you were at one time a field appraiser and now, for a lack of a better term, a review appraiser-meaning that you are reviewing reports in the hopes of increasing the company's QC. Back in the 90s I was a staff appraiser for BofA in southern California. During my employment 95% of my work was in the field. I rarely did review work but on occasion I had to. Every time I did review, I thought to myself it should be mandatory for all appraisers to do a bit of both. Reviewing is an eye-opener; and yet reviewers should go out into the field so they can remind themselves that field appraisers are not saving the "better" comps for another report.

Having said this I find that almost all adjustments-small or large-come mostly from Zoltan. Does an appraiser really do some type of paired sales for a half bathroom adjustment? I know I'm guilty of not doing a pair sales analysis for the majority of my adjustments. And what about the fields where no adjustment was made? The decision not to make an adjustment may also be from Zoltan-because I bet it didn't come from a pair sales analysis. I don't know.

What do you think?

 

[NLCApprMgr]

What do you think?

Due to profanity laws in the forum, I cannot say what I really think.

Market extraction, as a theory, is all well and good, but the real estate market is not perfect by far, and as far as I know, there is no current reliable data depository from which to extract the data or expedient software program to run the analysis.

I always went to the market. I called a few brokers with listings in the neighborhood, and I asked "how much more would a typical buyer pay for a second full bath as opposed to one and one half baths". Or...I would ask whether an S & M dungeon in a basement (yep, that happened once and not the strangest thing I ran into when I was in the field) added value, had no value, or detracted from value. While you might get three different answers, they were generally in the same ballpark. And that is what I told the report reader.

It's a heck of a lot better than running a random generator or searching the web for where Zoltan might be that day.

 

[Renee Healion]

I ask those same sorts of questions and I often ask what something cost. I know I can cost it out myself but often the respondent, in considering their answer, leads off naturally into a discussion of what a buyer would pay in light of what it would cost new. This is the same sort of calculation that in total informs a buyer's offer strategy which we reflect in our analysis and reporting.

 

[ucbruin]

Due to profanity laws in the forum, I cannot say what I really think.

Market extraction, as a theory, is all well and good, but the real estate market is not perfect by far, and as far as I know, there is no current reliable data depository from which to extract the data or expedient software program to run the analysis.

I always went to the market. I called a few brokers with listings in the neighborhood, and I asked "how much more would a typical buyer pay for a second full bath as opposed to one and one half baths". Or...I would ask whether an S & M dungeon in a basement (yep, that happened once and not the strangest thing I ran into when I was in the field) added value, had no value, or detracted from value. While you might get three different answers, they were generally in the same ballpark. And that is what I told the report reader.

It's a heck of a lot better than running a random generator or searching the web for where Zoltan might be that day.

I guess you make my point. Realtor = Zoltan in many cases. I applaud you for taking the extra step in your research but you still basically picked an adjustment value from out of thin air. Rationalize all you want, but that's what you did. You just included verbiage in your report that adjustments were based on Real Estate agents vs Zoltan in hopes of CYA. But I bet if you asked most residential appraisers they will agree that their adjustments for half bathrooms are from Zoltan and the overall "experience" and "knowledge" of the marketplace. Not a unreasonable thing to do.

 

[J Grant]

The fact that individual line item adjustments are a bit "fuzzy based", no matter how derived, is mitigated by practice...most of the time, it is not that critical whether a half bath is adjusted at $3000, or $5000. (or whether it was adjusted for at all)

Here is how to mitigate the fact that adjustments are not able to be precisely determined ( useful both in appraisal and reviewing)

1).Refer back to the unadjusted prices of the comps. Appraisers are not limited to basing their value opinions on the adjusted range of prices. The unadjusted range of prices are a "gold standard " as a value check. The unadjusted prices of the comps are "pure" prices, not tampered with due to adjustments. Thus, refer back to the unadjusted prices for a balanced view.

2). There hopefully is, one comp, or two ,, with few adjustments, or at least fewer critical adjustments. Often this comp or comps can be relied on toward reconciliation.

3). The appraiser bases their value opinion on more than just the adjusted sale prices. Appraisers who do lending work, read the statement above the value opinion on a URAR form, it sums up how the value was derived... references the inspection of subject, and entire SOW...in other words, other factors enter into reconciliation besides adjusted prices.

4)Reference back to a rough cost of the line item adjusted for. ( if it is a cost based item). Though cost does not equal value and some features can be over improvements, a rough idea of cost and the experience of market reaction relative to cost does helps define outer limits or reasonableness of an adjustment.

A $5000 full bath adjustment may be appropriate, in a 180k range house, ( less applicable in a $3 million luxury home where a bath can cost 40-60k...maybe a 20-30k k bath adjustment is more relevant. The market value definition references an informed buyer, which means they should have some rough idea of what things cost....so if items can be replaced or built for a certain cost, it points out adjustments far out of whack with costs relative to property type.

I am less concerned with small line item features, and more concerned with "big ticket" items, or those not cost based, such as views, or condo floor height, or location, which can have large impact on value.

 

[NLCApprMgr]

ucbruin -

I disagree with your comment. Realtors are in the market day in and day out. Good ones know their market. Good appraisers familiar with a market know which are the good realtors. If I only got one opinion, then the validity of the adjustment would be more questionable, which I why I asked a few. Like I said, usually, the realtors were pretty close in their opinions, which lent more credence to the adjustment I made.

If you have a different method, and you disclose that to the reader, then you are far ahead of what I see. Most of the time, the reports I read provide no evidence as to how the adjustment was quantified.

 

[J Grant]

NLCApprMgr,

Good point re Realtors.

A good way to "test" the reasonableness of adjustments is to arrive at them by different methods, and see how the results line up. Ask RE agents. Of course, with market experience, that is already done and may just need refreshing. Do paired sales on the grid, adjust for other features, see what is left and extract out value for line item.

Do a rough cost estimate, look at how new or depreciated the feature is, and see how it relates then to what a buyer might pay relative to quality/price range.

Thus, if you do paired sales and extract a range of 8k-12k for baths,, and agents state buyers pay 10-15 k for baths, and it costs 18-20k to build a new bath or remodel,, and the home is depreciated, say 14 years old, then a bath adjustment between 10 and 12k might be the most supportable as the various methods point to that range and each act as support for the adjustment.

 

[Renee Healion]

We do not just ask busy agents but homeowners and contractors and they have each done some measure of homework on these matters.

Refi times are great for retrofit and addition data. Building files contain project cost data. (HOs may under-report costs because the permit fee is tied to the project cost here, but when contractors pull permits on behalf of owners the figures are more likely to be accurate). These figures can be used when considering properties which then sell with such new/er work.

Investor rehabs offer clear lessons in contributory value.

It may seem merely anecdotal or scattershot. But the collection of various bits of information which we check against other information is what forms our body of historical knowledge, and which distinguishes what we offer from an AVM. It is the difference between the Universally Accepted Adjustment, applied everywhere, all the time, no matter the market, and the conclusions of someone who makes the effort to be a local expert.

In many cases, however, the work is not done to check that the conclusions are up to date. It really requires frequent re-examination to be sure you are not being complacent.