Legacy appraisal practice “the valuation process” is well established. Therein lies the problem. It is not designed to use modern methods toward objective, unbiased results.
First, lets distinguish the different kinds of bias. We need to separate personal bias from analytic bias.
Personal bias can be intentional, or unintentional (unconscious). Analytic bias itself can come from two parts of the valuation model: 1) bias in the data selection; and 2) bias in the model/algorithm selection.
Here we concern ourselves only with data selection. (However, note that data selection bias can be of the two types above: personal and model. Also, we look at only the traditional legacy appraisal model. Not AVMs or other hybrids or “non-appraiser appraisals.”
This traditional appraisal model follows a simple universal valuation path: 1) identify the problem; 2) select the comparable sales (the data); 3) apply adjustments; and 4) explain/reconcile inconsistencies.
In selecting three to six comps, we are told that they should be competitive, similar, and ‘able to be compared.’ To the best of my knowledge, no ‘similarity’ algorithms are to be found anywhere in the appraisal education requirements for licensing, nor in more ‘advanced’ text.
The selection of comparable sales is based on the appraiser’s knowledge of the location, the type of property, and features of the property (such as living area, site size, floor plan, or income potential). Traditionally, appraisers used three to six sales because:
- Sale data was incomplete and measurements seldom uniform from one property to another.
- Just gathering and verifying the data was a lot of work, perhaps 80% of the overall work.
- The human brain does not do well in understanding beyond six or seven data points.
Data bias comes from two sources: randomness of convenience — the ease of getting a sale data; and personal judgment bias of what is ‘truly’ comparable or not.
The motivation, or ‘source’ of the bias usually comes from the source of the work: the client. To keep a client, whether a lender, or management company (AMC) becomes the overriding motivation (conscious or unconscious). The unconscious part itself has several sources. The most obvious of these is called “anchoring.” Essentially, the human brain simply comes near the first number seen. (Like the sale price.)
As it turns out, this itself is justified by current subjective appraisal standards (USPAP), which require the appraiser to meet or exceed “the expectations of parties who are regularly intended users for similar assignments. This is followed by what I call Billie’s mom’s rule: “But mom! Billie’s mom lets him do it!”
Much of the current debate about how to modernize appraisal goes back to pervasive, root causes. Our expectations, habits, education, groupthink, and organizational policies and laws – all line up squarely to avoid looking at the two real issues: 1) lack of objectivity in data selection; and, 2) misuse of the very definition of ‘market’ value. (Market price does not equal value – see “loan conditions” adjustments.)
The FHFA and other governmental regulatory agencies must look squarely at the societal burden lined up against innovation and improvement of collateral evaluation systems. The burden of 55 state/territory/DC + Appraisal Foundation — fees and approvals inhibit progressive appraiser education. And prohibit education of technology-optimized valuation and risk assessment methods.
For appraisers, “getting CE hours for license renewal” has become detached from getting real progressive modern analytics education, which leads to unbiased results.
March 17, 2021 @ 7:23 am
As is nearly always the case we are on the same page with this, or more accurately, I am on your page.
What I find somewhat ironic is that very nearly the entirety of the appraisal industry as it relates to residential lending has accepted a flawed interpretation of “Market Value” as defined on the 1004 form which states that values are to be made in terms of cash or cash equivalency. Market value and Cash Value are most often substantially different figures, yet I have never seen an adjustment on any appraisal, not even my own, wherein the difference in value of a property is adjusted between an all cash sale and a sale that has financing, whether it be favorable financing or otherwise. Nor have I seen the ubiquitous “bracketing” requirement of all physical attributes by lenders be applied to “financing” terms and conditions. In this way, I would posit that only a fraction of appraisals performed actually satisfy the embedded definition of Market Value as a value in terms of cash or cash equivalency, and those that do meet the definition were not done that way purposefully, but rather their value conclusion equating to cash value is an unintended consequence of using all cash transactions in the analysis or some all cash transactions that happen to have some reasonable semblance of transactional value to the financed transactions within the report.
To that end….. Why not require one or more sales used in a report to be all cash? This would at least create some recognition of the potential value differences between a financed property wherein the buyer’s accepted risk is equivalent and even “scaled” to the amount of equity or down payment made. In the realm of FHA, we are talking as low as 3% or for VA 0% (and they deserve that consideration for their service). Whereas the cash buyer is 100% vested and any loss in value, no matter how small, is a personal loss.
This is why in a downward market the “all cash” transactions for homes is much lower than the financed transactions, because the cash buyer is able, as you pointed out in the article, to project that future conditions are expected to continue in a downward trend. Thus the cash buyer attempts to project the bottom of the market or what they perceive is a value that is acceptable based on future recovery expectations in the not too distant future. Example: offer $100k when projections show values continuing to decline to a $95k level but then will rebound and exceed the price paid at some short term in the future (accepting a temporary loss for a long term gain).
Conversely, in today’s rapidly appreciating markets like Seattle where I live, Cash is King and Cash Value sometimes exceeds what can be supported through traditional appraisal models which rely on dated sales, even if those sales are only 30-45 days old. But even in these situations, the Cash offer is less than the financed offers, but is accepted out of fear that the appraised value will not support the contract price or, to put it bluntly, the seller takes the least amount of risk in an all cash sale which exceeds initial expectations of a market value (financed) sale based on the list price. Or put another way, the seller accepts a “Less than Market Value” offer because it represents a lower risk of the escrow failing.
Conclusion: In what would be considered “normal expectations” of market participants (buyers/sellers) along the entire spectrum of “market conditions” (appreciating/stable/depreciating) Cash/Collateral value will nearly always be less than Market Value because in a rapidly appreciating market the seller typically will accept less risk by accepting a lower offer that has a better chance of closing and in a rapidly depreciating market the buyer will take less risk by offering a lower price. As with all things, there are and will be anomalous periods and markets to this conclusion, but that’s just me hedging my conclusion against the uncertainty of certainties.
Todd Redington, SRA AI-RRS
March 17, 2021 @ 12:20 pm
Statistics measure the central tendency of a population, which, by definition (in theory and by some interpretations) is “market value”. While analytics will render a single value point that represents the general thinking of the population as a whole, does it represent the true value of the subject property? By its very nature, the real market is somewhat bias, because each home is unique as the individuals selling or buying it. People have preferences in colors, décor, locations, style, amenities, features, etc., that will vary from one person to another. Small groups of people will have many of these in common, just like new homes from the builder are highly similar, however, after the newness wears off, it changes. On a resale basis, there are too many variables (the contribution to value of which cannot be accurately measured) between groups of homes and the individuals buying and selling them. Therefore, the impact of the variables cannot be quantified by analytics alone.
Statistically, we can arrive at an “unbiased” preliminary conclusion as to the value of a property based upon the actions of a large population. However, from that point, we must consider and account for the factors that are not accurately measured by statistics alone, in order to arrive at a final value conclusion. The “most probable price” applies to the subject property as of the date of value. While the central tendency of a large data set may represent the value of the typical 3-bedroom, 2-bath, 1500 SF home in a given market area, does it represent the value of the subject property?
Regression (like justice) is somewhat blind and does not consider all factors that impact the price paid by individuals for a given property. Using stats alone, a house that is purple inside and out, has the same market value as other homes in the same area with the same physical characteristics that may be more conforming to the market in color and décor. While analytics will produce an “unbiased” opinion of value of that purple house, will analytics alone, accurately measure “the most probable price” for the purple home?
As George will tell you, analytics is a tool to be employed by the appraiser to solve the appraisal problem, just like the approaches to value, or other interim steps and methods. Market value is not the result of “central tendency”. However, the central tendency of market participants is a component of the analysis that provides a baseline from which we can reconcile the final value opinion.
We are engaged to solve the appraisal problem and we are required to do this in an objective and unbiased manner. I think most would agree that analytics will help us in this process. However, analytics is not the be-all and end-all solution to the appraisal problem. This is why we have reconciliation in each of the approaches and in the final opinion of value. Reconciliation is perhaps the most important and overlooked part of the process.
Interesting article … just my two cents.
Patrick Egger
March 19, 2021 @ 10:23 am
George, I am reminded of my first year appraising and running into racial bias. Here is my experience:
my first year appraising was for a S&L that had been sued by the Feds for Redlining. There were 14 different areas that were UnderServed with loans.
Because I was the only trainee they had with any real estate education or espericence; they sent me into these formerly Redlined areas.
They had a huge prepayment penalty, so if they ever made a loan on a poroperty, it alwasy came back to them upon sale or refi. And, they would send the old appraisal up.
One day I was appraising in Altadena, CA. I had the old report and the only thing written into the Neighborhood comments was “evidence of Colored moving into the area”.
I was shocked, so I looked at who signed it. It was another trainee, in 1961. He now was tyhe Chief Appraiser, MAI, SRA, SRPA.
Mind you, i came after Redlining had been stopped by legal action, and after which they all reported loans made by Zip Code, so the Feds could keep track.
Discrimination was real!
Decades later, while still working in the Lost Angeles general area, I was doing research on real estate frauds, mortgage frauds and appraisal frauds.
HUD had a list of Zip Codes which were high default rate areas. They had them on a map, a colored map. The highest default rate areas were in Red, then Orange, then Yellow.
Interestingly, these areas almost mirrored the same areas Redlined in the 1960’s.
Low income areas, regardless of race, become targets for predatory lending. Some loans were so bad, that the loan killed the borrowers ability to continue paying.
“Real Estate Frauds and Appraisers Legal Liabilities and Responsibilities”, Real Estate Fraud’s and the Appraisers Role”, “Predatory Lending, Client Pressures and Appraisal Frauds”, “Mortgage Frauds and the Appraisers Role”, circa 1999-2003.
I delivered seminars across the country, trying to help appraisers understand how they were being used sometimes, duped other times; in an effort to help appraisers stay out of trouble.
Most Mortgage Fraud cases involve inflating the price and either getting a dirty appraiser, unaware appraiser or even forging an appraisal to make the deal work. The appraisers became the unwitting enablers to many mortgage frauds and predatory lending.
Back then, calls were being made to see if the appraiser could come up with the value needed to do the deal. Sometimes the whole transaction was a fraud, an illegal flip as an example, or predatory because they were packing 10-20 Points into the front end of the loan, feathering the pocket of the loan agent.
Every appraiser that searches for comps based on Sales Price, is subject to being duped. I was trained to Bracket the sales price, something I questioned on day one.
Bottom line, it has been my experience over a long career, that there is more evidence of appraisers inflating value, than deflating it, because that is where the pressures come.
How Do I Move to EBV? Part II - George Dell, SRA, MAI, ASA, CRE
May 24, 2023 @ 1:15 am
[…] generates more than a point value opinion. Therefore it requires competence beyond old legacy appraisal “procedures.” This new competency is a merger of traditional practice with modern data […]